functional analysis of the drosophila gene smallish (cg43427)

Functional analysis of the Drosophila

gene smallish (CG43427)

PhD Thesis

for the award of the degree

“doctor rerum naturalium”

in the GGNB program Genes and Development

at the Georg August University Göttingen,

Faculty of Biology

Submitted by

Seyed Amir Hamze Beati

born in Henstedt-Ulzburg, Germany

Göttingen, 2013

III

PhD Thesis Committee

Prof. Dr. Andreas Wodarz (Uni-Med)

Department of Anatomy and Cell Biology (Stem Cell Biology)

Georg-August-University, Göttingen

Prof. Dr. Ernst Wimmer (Uni-Bio)

Department of Developmental Biology

Georg-August-University, Göttingen

Prof. Dr. Reinhard Schuh (MPI-bpc)

Department of Molecular Developmental Biochemistry (Molecular Organogenesis)

Max Planck Institute for Biophysical Chemistry, Göttingen

Day of PhD examination: 13th march 2013

V

AFFIDAVIT

I hereby declare that I prepared the thesis “Functional analysis of the Drosophila gene smallish

(CG43427)” on my own with no other sources and aids than quoted.

Seyed Amir Hamze Beati

Göttingen, January 31st 2013

TABLE OF CONTENTS

VII

TABLE OF CONTENTS

Acknowledgments ............................................................................................................................. XI

Abstract ........................................................................................................................................... XIII

List of figures .................................................................................................................................... XV

List of tables ................................................................................................................................... XVII

1 Introduction .................................................................................................................................... 1

1.1 Cell polarity ........................................................................................................................... 1

1.2 Cell adhesion ......................................................................................................................... 6

1.3 Src kinases ............................................................................................................................. 9

1.4 LMO7 ................................................................................................................................... 16

1.5 smallish (CG43427).............................................................................................................. 21

1.6 Scope of the thesis .............................................................................................................. 24

2 Material and methods .................................................................................................................. 25

2.1 Chemicals and materials ..................................................................................................... 25

2.1.1 Chemicals and enzymes ................................................................................................ 25

2.1.2 Kit systems .................................................................................................................... 25

2.1.3 Photo and picture analysis ........................................................................................... 26

2.1.4 Bacterial strains and cell culture lines .......................................................................... 26

2.1.5 Plasmids ........................................................................................................................ 27

2.1.6 Buffers and medium ..................................................................................................... 27

2.1.7 Primers .......................................................................................................................... 32

2.1.8 Primary antibodies ........................................................................................................ 35

2.1.9 Fly stocks ....................................................................................................................... 36

2.1.10 Fly breeding .................................................................................................................. 38

2.2 Genetic methods ................................................................................................................. 38

2.2.1 Separation of DNA fragments via gel electrophoresis ................................................. 38

TABLE OF CONTENTS

VIII

2.2.2 Polymerase Chain Reaction (PCR) ................................................................................. 39

Taq polymerase 30 - 60 sec = 1 kb ................................................................................................... 40

Pfu polymerase 90 sec = 1 kb ........................................................................................................... 40

2.2.3 Mutagenesis PCR .......................................................................................................... 40

2.2.4 Transformation of DNA into chemically competent E.coli ........................................... 41

2.2.5 Isolation of DNA out of an agarose gel ......................................................................... 41

2.2.6 Purification of DNA out of a PCR product ..................................................................... 41

2.2.7 pENTRTM/D-TOPO® cloning ............................................................................................ 42

2.2.8 Isolation of DNA from bacteria ..................................................................................... 45

2.2.9 Restriction digestion ..................................................................................................... 45

2.2.10 Determining DNA concentration .................................................................................. 45

2.2.11 Sequencing of DNA ....................................................................................................... 46

2.2.12 Isolation of genomic DNA from flies ............................................................................. 46

2.2.13 UAS/Gal4 system .......................................................................................................... 47

2.2.14 Generation of transgenic flies....................................................................................... 48

2.3 Biochemical methods .......................................................................................................... 50

2.3.1 SDS-PAGE ...................................................................................................................... 50

2.3.2 Western blot ................................................................................................................. 51

2.3.3 Lysis of Drosophila embryos ......................................................................................... 51

2.3.4 Lysis of Drosophila S2 cells ........................................................................................... 52

2.3.5 Determination of protein concentration ...................................................................... 52

2.3.6 Co-Immunoprecipitation (Co-IP)................................................................................... 52

2.4 Cellculture ........................................................................................................................... 54

2.4.1 Transfection of S2 Schneider cells with FuGENE HD transfection reagent .................. 54

2.5 Histology .............................................................................................................................. 54

2.5.1 Formaldehyde fixation of embryos............................................................................... 54

TABLE OF CONTENTS

IX

2.5.2 Formaldehyde fixation of larval tissue ......................................................................... 55

2.5.3 Heat fixation of embryos .............................................................................................. 55

2.5.4 Immunofluorescence staining ...................................................................................... 55

2.5.5 Cuticle preparation ....................................................................................................... 56

2.5.6 Wing preparation .......................................................................................................... 56

2.5.7 Eye preparation ............................................................................................................ 56

3 Results ........................................................................................................................................... 58

3.1 Baz binds to Smash in vivo .................................................................................................. 58

3.2 Expression pattern of smash ............................................................................................... 59

3.3 smallish knockout ................................................................................................................ 66

3.4 Overexpression of Smash .................................................................................................... 74

3.5 Smash binds Src42A and Src64B in vivo .............................................................................. 83

3.6 Overexpression of Src42A in the eye .................................................................................. 91

3.7 smallish genetically interacts with Src64B .......................................................................... 93

4 Discussion...................................................................................................................................... 97

4.1 Baz binds to Smash in vivo .................................................................................................. 97

4.2 Expression of smallish ......................................................................................................... 99

4.3 Knockout of smallish ......................................................................................................... 100

4.4 Overexpression of Smash .................................................................................................. 101

4.5 Smash forms a complex with Src42A in vivo and interacts genetically with Src64B ........ 102

5 Conclusion and future perspectives ........................................................................................... 105

6 References ................................................................................................................................. CVIII

ACKNOWLEDGEMENTS

XI

Acknowledgments

Firstly, I would like to thank Prof. Dr. Andreas Wodarz for giving me the opportunity to do my PhD

thesis in his lab. I really appreciate the helpful discussions, ideas, support and that I had the

possibility to develop my own ideas. Thank you.

I further would like to thank Prof. Dr. Ernst Wimmer and Prof. Dr. Reinhard Schuh for being part of

my thesis committee and for the helpful discussions and suggestions during my committee

meetings.

In addition I would like to thank the GGNB program “Genes and Development” for organization,

helpful discussions and the nice atmosphere at the retreats and the GZMB, CMPB and DFG for

funding this project.

I am grateful to Shigeo Hayashi for providing pSrc antibody, Kaoru Saigo for the Src42A antibody

and Src42A mutants and Michael A. Simon for providing Src64B mutants.

Meinen besonderen Dank möchte ich allen der Abteilung Stammzellbiologie widmen. Es war eine

sehr schöne Zeit in der ich viel gelernt habe und vorallem auch sehr viel Spaß gehabt habe. Mona,

ich weiß du magst nicht vorne stehen, aber vielen Dank für deine stetige Hilfsbereitschaft und das

Klonieren von unklonierbaren Konstrukten (ich bin mir sicher selbst Titin wäre vor dir nicht

sicher!). Karen, dir danke ich für das gegenseitige Aufbauen während unserer beider Arbeiten,

dein stets offenes Ohr und vorallem für deine Freundschaft. Vielen Dank dafür! Jaffer möchte ich

für das coolste „Fellowship“ des Universums danken. Ganz besonderen Dank auch für die vielen

Korrekturen. Thanks Bro! Oh Carolina, dir danke ich natürlich auch für die zahlreichen lustigen

Abende mit unserem Kumpel Jack und dem stetigem Versuch das Ameisensterben zu stoppen.

Sascha, auch wenn du schon zu den „Ehemaligen“ gehörst danke ich dir für eine coole Zeit im

Labor, auch wenn dein Musikgeschmack echt grenzwertig ist… Bei Katja (Krusty) möchte ich mich

gerne für den „Motivationsschub“ in der Endphase dieser Arbeit bedanken! Patricia, vielen Dank

für den ganzen bürokratischen Teil und mein erstes Whisky Seminar. Das wird definitiv sehr bald

wiederholt! Ich möchte mich weiterhin bei Katja (Curthi) und Claudia für die stetige Hilfe im Labor

und die jährlichen BBQs bedanken, sowie Julia für das super Fliegenfutter! Manu danke ich für die

Diskussionen und Ideen. Natürlich gilt besonderer Dank auch Ieva, Marilena, Gang, Nils und Tobi.

Ihr seit mir alle sehr ans Herz gewachsen und werde die gemeinsame Zeit mit euch nicht

ACKNOWLEDGEMENTS

XII

vergessen. Ich werde diese als eine meiner positivsten in Erinnerung behalten. Mein weiterer

Dank gilt auch den Molekularen Onkologen für die gemeinsame Zeit auf unserem Flur, den netten

Gesprächen bei der Kaffeerunde und den zahlreichen Aktivitäten außerhalb des Labors.

Mein besonderer Dank gilt meiner Familie und meinen Freunden. Ihr habt mir immer den Rücken

gestärkt, mich unterstützt und motiviert sobald es nötig war. Dies hat mir letztlich sehr geholfen

alle Tiefs zu durchstehen. Vielen, vielen Dank!

ABSTRACT

XIII

Abstract

The establishment and maintenance of cell polarity is crucial for the function of many cell types in

multicellular organisms. Especially in epithelial tissues, cell polarity is connected to the regulation

of cell adhesion and regulated by a complex hierarchy of highly conserved proteins. These can be

subdivided into three groups of genes, the bazooka and crumbs groups, which encode apically

localizing proteins, and the discs large group that encodes laterally localizing tumor suppressor

proteins. Among these classes of proteins, Bazooka (Baz), the Drosophila homolog of vertebrate

Par-3, plays a predominant role as shown by genetic epistasis experiments.

In a yeast two-hybrid screen we identified the protein encoded by the annotated gene CG43427

which we named smallish (smash), as a new interaction partner of Baz. The gene product of

smash possesses a C-terminal PDZ binding motif and a LIM domain close to the C-terminus.

Endogenous Smash colocalizes with Baz apically in epithelial cells, a region harboring the

adherens junctions (AJs). Co-immunoprecipitation of Baz and an N-terminally tagged version of

Smash-PI (an isoform encoding for the last 889 amino acids of Smash) has confirmed that these

proteins interact in vivo in embryos.

To analyze the function of smash during the development of Drosophila, we generated two

different knockout alleles by transdeletion, one representing a null allele and the other a C-

terminal truncation affecting the part of the protein carrying the LIM domain and the PDZ binding

motif. We found that smash is not an essential gene, as homozygous mutants for both alleles are

viable and fertile. The subcellular localization of polarity markers such as Baz were not affected

upon smash knockout. On the other hand, overexpression of Smash using the UAS/Gal4 system

and transgenes encoding for N-terminally GFP-tagged versions of Smash caused lethality in

embryonic and larval stages. Rare eclosing escaper flies were decreased in body size.

Overexpression of Smash in epithelial cells resulted in reduction of the apical surface area,

indicating that Smash may function in apical constriction, a process important for morphogenetic

rearrangements in epithelia. Overexpression of Smash during eye development caused a rough

eye phenotype and reduction of eye size. Upon ubiquitous overexpression of Smash in embryos,

many embryonic cuticles exhibited anterior and dorsal holes.

ABSTRACT

XIV

Following up on these findings, we showed that the non-receptor tyrosine kinase Src42A binds N-

terminally GFP-tagged versions of Smash-PI in vitro in S2 cells and is furthermore able to

phosphorylate GFP-Smash-PI. Endogenous Smash protein was found to be tyrosine

phosphorylated in vivo in embryos as well. Domain deletion versions of Src42A still showed

binding to Smash, indicating different binding mechanisims provided by the fact that tyrosine

phosphorylation of Smash was only abolished upon deletion of the kinase domain.

A double mutant for Src64B, the second Src kinase encoded by the Drosophila genome, and

smash is lethal. However, embryonic cuticles did not show defects and epithelial integrity

appeared intact.

LIST OF FIGURES

XV

List of figures

Fig.1: Organization of epithelial cells in comparison between Drosophila and vertebrates ............. 2

Fig.2: Localization of distinctive protein markers in Drosophila epithelial cells ................................ 5

Fig.3: Classical model of cell-cell adhesion ........................................................................................ 7

Fig.4: Revised models showing possibilities of the linkage between AJs and the cytoskeleton ........ 8

Fig.5: Structure of Src kinases .......................................................................................................... 10

Fig. 6: Model of dorsal trunk elongation with regard to Src42A function ....................................... 13

Fig.7: Dosal closure and JNK signaling .............................................................................................. 15

Fig.8: Localization of LMO7 in epithelial cells of rat gallbladder...................................................... 18

Fig.9: LMO7 structure and function ................................................................................................. 21

Fig.10: Principle of Gateway® cloning .............................................................................................. 42

Fig.11: UAS/Gal4 system .................................................................................................................. 48

Fig.12: Yeast two-hybrid screen with PDZ domains of Baz as bait................................................... 58

Fig.13: Smash colocalizes with Baz and binds to in vivo .................................................................. 59

Fig.14: Embryonic expression pattern of smash .............................................................................. 62

Fig.15: Subcellular localization of Smash ......................................................................................... 64

Fig.16: Subcellular localization of Smash transgenes ....................................................................... 66

Fig.17: Transdeletion of the genomic locus of smallish ................................................................... 67

Fig.18: Verification of smash knockout ............................................................................................ 68

Fig.19: Lethalities shown for smallish knockout .............................................................................. 70

Fig.20: Epithelial integrity is not lost upon smash knockout ........................................................... 72

Fig.21: Protein levels of AJ and polarity markers in smash mutants ............................................... 74

Fig.22: Lethality after overexpression of GFP-Smash epitopes ....................................................... 75

Fig.23: Size decrease upon GFP-Smash-PI expression ..................................................................... 76

Fig.24: Cuticle phenotypes observed after overexpression of GFP-Smash-PM ............................... 78

Fig.25: Overexpression of GFP-Smash-PM leads to cells smaller in size .......................................... 79

LIST OF FIGURES

XVI

Fig.26: AJ length and apical surface area is reduced upon GFP-Smash-PM expression .................. 80

Fig.27: Expression of GFP-Smash-PM does not show an effect on wing shape ............................... 81

Fig.28: Eye restricted expression of Smash leads to rough eyes and size reduction ....................... 82

Fig.29: Subcellular localization of Src42A and the activated form pSrc ........................................... 85

Fig.30: GFP-Smash binds to Srcs and is tyrosine phosphorylated in vivo ........................................ 87

Fig.31: Src deletion Co-IPs ................................................................................................................ 88

Fig.32: Analysis of Smash-PI phosphomutants ................................................................................. 90

Fig.33: Expression of Src in the eye .................................................................................................. 93

Fig.34: Lethality of smash and Src64B double knockout .................................................................. 94

Fig.35: Double-knockout of smash and Src64B shows no dorsal closure defects and normal

epithelial integrity ............................................................................................................................ 96

LIST OF TABLES

XVII

List of tables

Table 1: Used bacterial strains and cell lines ................................................................................... 26

Table 2: Used plasmids and their properties ................................................................................... 27

Table 3: Used primers and their application .................................................................................... 33

Table 4: Primary antibodies ............................................................................................................. 35

Table 5: antibodies and fluorochrome conjugated phalloidin ......................................................... 35

Table 6: Fly stocks ............................................................................................................................ 36

Table 7: Example for PCR program .................................................................................................. 40

Table 8: Constructs generated in this work ..................................................................................... 43

Table 9: Sequencing program .......................................................................................................... 46

Table 10: Phenotypical markers for identification of the chromosome carrying the transgene ..... 49

Table 11: Examples of contents for different molecular SDS-PAGE gels ......................................... 50

Table 12: Affinities of protein A/G agarose for different immunoglobulins .................................... 53

INTRODUCTION

1

1 Introduction

1.1 Cell polarity

The polarization of a cell regulates various aspects of cell behaviour, such as the shape of the cell,

the unequal distribution of organelles or the alignment of the cytoskeleton. Cell membranes are

furthermore composed of different types of lipids, which also represents a type of polarization. A

very important feature of polarization is provided by the asymmetric localization of different

proteins or protein complexes. Many types of cells are polarized, e.g. neurons, oocytes or stem

cells, to mention a few.

Epithelial cells represent a highly polarized cell type and have important functions in forming

physiological and mechanical barriers (Suzuki and Ohno, 2006) and in shaping a metazoan

organism by delineating different compartments (Knust and Bossinger, 2002). The plasma

membrane of epithelial cells can be subdivided into two distinct domains: the apical membrane

domain facing the environment or a lumen and the basolateral membrane domain, which is in

contact with neighboring cells and the basal substratum. These two membrane domains are

segregated by highly elaborated adherens junctions (AJs). Fig.1 shows a schematic of an

ectodermal epithelial cell of Drosophila melanogaster (Drosophila) in comparison with an

ectodermal epithelial cell of vertebrates.

The region containing the AJs is also referred to as the zonula adherens (ZA). A region slightly

above the ZA is called marginal zone or subapical region (SAR), which harbours proteins which

have been identified as homologs of vertebrate tight junction (TJ) proteins. However, TJs are

absent in Drosophila which in contrast features septate junctions (SJ) at the lateral membrane,

which are not formed in vertebrates (Fig.1). Within these membrane domains three main protein

complexes had been identified over the past two decades, which localize in a highly polarized

fashion to these distinct regions. These complexes will be discussed in more detail in the following

pages with regard to their function in ectodermal epithelia.

As mentioned above, another highly polarized cell type is represented by stem cells. The

Drosophila ventral neural ectoderm (VNE) is the origin for Drosophila neuroblasts (NB), which will

give rise to the nervous system of the animal. Here, NBs divide asymmetrically, which leads to the

generation of two daughter cells, another NB and a ganglion mother cell (GMC). The latter cell will

INTRODUCTION

2

divide once more and give rise to a pair of neuron or glial cells, whereas the NB will continue

dividing (Wodarz and Huttner, 2003; Wodarz, 2005). In the VNE, individual neuroectodermal cells

are determined to become NBs via Notch/Delta signaling and delaminate from the epithelium

into the embryo (Doe, 2008). A very important point is that the NB will inherit the polarization of

the neuroectodermal cells, which indicates that those proteins needed for polarization are not

just important for epithelial cells but also for other types of cells.

Fig.1: Organization of epithelial cells in comparison between Drosophila and vertebrates

(A) Epithelial cells of Drosophila can be distinct into different regions: an apical membrane domain facing

the environment or a lumen, and a basolateral membrane domain which is in contact with neighboring cells

as well as with the basal substratum. Both domains are segregated by AJs, which is a belt-like structure

encircling the cell, also referred to as ZA. Apical to the ZA a region is defined as SAR or marginal zone. The

latter region harbours protein homologs of vertebrates which form TJs (B). Although TJs are not formed in

Drosophila, proteins localizing to this region share some functions. In comparison Drosophila exhibits SJs,

which are absent in vertebrates. Adapted from Knust and Bossinger, 2002.

As mentioned above epithelial cells are highly polarized and depend on three identified groups of

proteins or genes which are involved in the correct formation and maintenance of epithelial

integrity. The gene products of two of the three groups were found to be localized apically in

epithelial cells, regions referred to as the apical membrane domain and the ZA. Gene products of

the third group have been shown to localize to the lateral membrane domain and the SJs. The

apical protein complexes belong to gene products of the bazooka (baz) and the crumbs (crb)

group. The discs large (dlg) group represents proteins of tumor suppressor genes, which have

been found to localize at the lateral membrane and the SJs. These three groups have been shown

INTRODUCTION

3

to be crucial for the establishment of epithelial cell polarity as well as their maintenance and will

be discussed in more detail below (Johnson and Wodarz, 2003).

Proteins belonging to the baz group of genes are Baz, which is the Drosophila homolog of

vertebrate Partitioning defective 3 (Par3), Drosophila atypical Protein Kinase C (DaPKC) and the

adaptor protein Drosophila Partitioning defective 6 (DPar6). Baz and DPar6 are scaffolding

proteins, which exhibit PDZ domains (name derived from PSD-95, Dlg and ZO-1). PDZ domains are

one of the most common protein-protein binding domains (Sheng and Sala, 2001; Te Velthuis and

Bagowski, 2007). They consist of about 80-90 amino acids, which contain six anti-parallel β-

strands and two α-helices (Fanning and Anderson, 1999). They bind to C-terminal peptide motifs

and internal sequences resembling a C-terminus and are also described to bind phospholipids

(Harris and Lim, 2001; Jeleń et al., 2003). Baz, DaPKC and DPar6 are also referred to as the Par

complex, since they had been found to form a protein complex in vivo (Wodarz et al., 2000;

Petronczki and Knoblich, 2001). The binding of Baz to DPar6 and DaPKC is important for their

initial recruitment to the apical plasma membrane (Harris and Peifer, 2005; Horikoshi et al., 2009).

Later DaPKC phosphorylates Baz at serine 980 and thereby releases it from the complex. DaPKC

and DPar6 remain in the SAR due to the binding of DPar6 to Crb (Morais-de-Sá et al., 2010;

Walther and Pichaud, 2010), whereas Baz localizes to the AJs (Nam and Choi, 2003; Harris and

Peifer, 2005; Horikoshi et al., 2009; McCaffrey and Macara, 2009; Morais-de-Sá et al., 2010;

Walther and Pichaud, 2010). DPar6 acts as a regulatory subunit of DaPKC with evidence showing

that it negatively influences its kinase activity (Atwood et al., 2007), which is of importance for the

maintenance of apical membrane identity. For example phosphorylation of Lethal giant larvae

(Lgl, which is a member of the dlg group) and Par1 leads to their exclusion from the apical

membrane (Betschinger et al., 2003; Plant et al., 2003; Yamanaka et al., 2003; Hurov et al., 2004;

Kusakabe and Nishida, 2004; Suzuki et al., 2004). However, it was found that DaPKC

phosphorylates the cytoplasmic tail of Crb at four serine/threonine residues (Sotillos et al., 2004),

but the in vivo function of this modification remains unknown (Huang et al., 2009). As mentioned

above, phosphorylation of Baz results in its dissociation from the Par complex and relocalization

to the AJs. Here Baz can bind to Armadillo (Arm, the Drosophila homolog of β-Catenin (β-Cat)) and

Echinoid (an immunoglobulin-superfamily adhesion molecule) (Wei et al., 2005) and to a

phosphatase PTEN (Von Stein et al., 2005). Here Baz has been proposed to function in the

recruitment of cadherin-catenin clusters for the formation of AJs (McGill et al., 2009). With

regards to this, baz loss of function alleles result in a loss of AJs components and the phenotype

resembles the loss of function of arm (Müller and Wieschaus, 1996). Furthermore, apical polarity

INTRODUCTION

4

markers are reduced and were found to be mislocalized along the basolateral membrane domain.

Cells are rounded up and the epithelium becomes multilayered. As a consequence these cells

begin to die through apoptosis (Bilder et al., 2003).

The crb group, which is the second group of proteins localizing to the apical plasma membrane

domain, consists of Crb, which is the only transmembrane protein (among the so far identified

members of the three groups) with a huge extracellular domain consisting of EGF and LamG

domains. It exhibits a short intracellular tail of 37 amino acids containing a highly conserved C-

terminal PDZ binding motif (ERLI), which recruits Stardust (Sdt, encoding for a membrane

associated guanylate kinase (MAGUK)) as a member of the crb group to the apical membrane.

Aside from a single PDZ domain, Sdt exhibits a L27 and a SH3 domain and recruits PATJ (Pals1-

associated TJ protein) to the apical membrane, which also contains a L27 domain as well as four

PDZ domains. Crb is localized slightly apical to the AJs in the SAR (Tepass, 1996) and crb mutants

show loss of apical membrane identity and the AJs, whereas overexpression leads to an increase

of the apical membrane domain (Wodarz et al., 1993, 1995).

The dlg group of tumor suppressor genes is composed of Dlg and Scribble (Scrib), which exhibit

several PDZ domains, as well as Lgl, a WD40 domain containing protein. These polarity markers

are located at the lateral plasma membrane. Scrib was also described to exist in a cytoplasmic

pool (Bilder and Perrimon, 2000; Bilder et al., 2000). In contrast to proteins of the apical

networks, members of the dlg group have not been described to bind to each other. Mutations in

these genes show abnormal cell shapes and loss of the ZA accompanied by a multilayered

epithelium (Bilder and Perrimon, 2000; Bilder et al., 2000, 2003; Tanentzapf and Tepass, 2003). A

very important difference to mutations in the baz and crb group is an enlarged apical membrane

domain, which is reduced or lost in mutations of the latter genes. Furthermore mutations in genes

of the dlg group do not lead to apoptosis of these cells (Bilder and Perrimon, 2000; Bilder et al.,

2000, 2003). Fig.2 shows a schematic of an epithelial cell with the main identified polarity markers

which play a role in establishing or maintaining epithelial polarity and integrity.

INTRODUCTION

5

Fig.2: Localization of distinctive protein markers in Drosophila epithelial cells

Different distinguishable regions of the epithelium are indicated on the left side of the scheme. Members of

the baz and crb group are shown in orange and those of the dlg group are shown in blue. Baz was also

identified to be an AJ marker, where it recruits cadherin-catenin clusters (CCC). Arrows indicate the

interaction between AJ markers and apical polarity determinants, negative regulatory mechanisms are

indicated between proteins of the lateral membrane domain and apical polarity proteins. Adapted from

Tepass, 2012.

Genetic experiments revealed that baz gene function is most likely upstream of other identified

genes that encode for polarity markers so far. As mutations in crb or sdt, as well as baz, show

quite similar phenotypes, defects in baz mutants become apparent slightly earlier. Furthermore

Crb mislocalizes in baz mutants, but Baz is localized correctly in crb mutants (Müller and

Wieschaus, 1996; Müller, 2000; Bilder et al., 2003). In this context it was shown that Baz recruits

Sdt to the plasma membrane. This direct interaction is dependent on aPKC activity, as

phosphorylation of Baz at serine 980 causes dissociation of Sdt from the complex. Expression of a

respective non-phosphorylatable Baz transgene caused phenotypes similar to crb and sdt mutants

(Krahn et al., 2010a). It has been shown that proteins of these complexes interact in a dynamic

manner (some examples had been discussed above). One important regulatory mechanism was

identified by genetic experiments, where it was found that apical determinants antagonize the

function of laterally localized proteins, and vice versa. For example zygotic crb scrib double

mutants somehow show suppression of the crb single mutant phenotype to a large extent,

INTRODUCTION

6

indicating the interaction between these two different groups (Bilder et al., 2003). However,

zygotic dlg baz double mutants show quite a similar phenotype to baz single mutants, underlining

the epistatic importance of baz in the establishment of cell polarity (Bilder et al., 2003; Tanentzapf

and Tepass, 2003).

There are many more factors which are important for the establishment and maintenance of

epithelial polarity and integrity. For example Yurt (Yrt), Coracle (Cora), the NaK-ATPase and NrxIV

have been shown to be necessary for proper SJ formation and are implicated in tube size control

of tracheal cells, which also represent a type of an ectodermal epithelium (Laprise et al., 2010).

Since Echinoid was recently shown to fuction upstream of the Hippo pathway (Yue et al., 2012),

which in general is a pathway described for being important for tissue growth and organ size

(Cherret et al., 2012), the establishment of cell polarity and junction formation must be regarded

as a highly dynamic process.

1.2 Cell adhesion

AJs, which have already been mentioned in the previous chapter, are important for cell-cell

adhesion. They are composed of E-Cadherin (E-Cad), a transmembrane protein which is important

for the homophilic cell-cell adhesion and its intracellular associated Catenins. The extracellular

domain of E-Cad forms trans dimers with E-Cad proteins of the plasma membrane of the

neighboring cell and cis dimers with E-Cad proteins of the same cell. Intracellular, E-Cad binds to

β-Cat which in turn binds to α-Catenin (α-Cat). The complex of E-Cad-β-Cat-α-Cat is also referred

to as cadherin-catenin complex (see chapter before). α-Cat associates with Actin and it was

believed that these interactions form a stable link between AJs and the cytoskeleton. Nelson and

co-workers have shown in 2005 that the function of AJs in cell adhesion is much more dynamic

and that a quaternary complex of E-Cad-β-Cat-α-Cat-Actin cannot exist simultaneously (Drees et

al., 2005; Gates and Peifer, 2005; Yamada et al., 2005). It was shown that a monomeric form of α-

Cat binds to the E-Cad-β-Cat complex, whereas a dimeric form of α-Cat does not bind to this

complex anymore. In contrast, these homodimers show high binding affinity for Actin.

Furthermore α-Cat homodimers can suppress the activity of the Arp2/3 complex, which is

important for the nucleation of Actin branches. However, the physiological relevance of this

property is not known. Fig.3 shows the classical view of cell-cell adhesion as described above and

INTRODUCTION

7

Fig.4 shows an illustration of the revised model of how AJs may be connected tightly or transiently

to the cytoskeleton.

Fig.3: Classical model of cell-cell adhesion

(A) AJs are important for cell-cell adhesion. Here, E-Cad as a transmembrane protein forms cis and trans

dimers. The latter are important for cell-cell contacts. (B) Intracellularly, E-Cad associates with β-Cat, which

in turn binds to α-Cat. Since α-Cat binds to Actin it was believed that this binding forms a stable link

between AJs and the cytoskeleton. Other Actin binding proteins like ZO-1 or Afadin have been proposed to

play a role in this link as well, as many of them can associate with α-Cat. Adapted from Gates and Peifer,

2005.

INTRODUCTION

8

Fig.4: Revised models showing possibilities of the linkage between AJs and the cytoskeleton

(A) Summarized revised interactions of the cadherin-catenin complex. α-Cat associates with E-Cad-β-Cat in

a monomeric form, whereas its dimeric form dissociates from this complex and shows binding to Actin.

However, homodimers of α-Cat can antagonize the activity of the Arp2/3 complex. (B) A possible link

between AJs and the cytoskeleton could be through Nectins. These transmembrane proteins belong to the

immunoglobulin-superfamily adhesion molecules. Nectins form dimers as well and associate with Afadin

intracellularly, which in turn binds to Actin, providing a second link between AJs and the cytoskeleton. (C) A

more complex model includes many protein-protein interactions which thereby form a transient link

between AJs and the cytoskeleton, which is highly dynamic. Adapted from Gates and Peifer, 2005.

The remodelling and interplay of AJs and the Actin cytoskeleton is of fundamental importance, as

processes like apical constriction, where the Actin/Myosin ring beneath the AJs contracts to

mediate cell shape changes, are needed for morphogenetic processes. e.g., during gastrulation

the mesoderm invaginates due to repositioning of the AJs by contraction forced by the

Actin/Myosin network. When AJ function is abolished by depletion of β-Cat, the Actin/Myosin ring

still contracts, but cell shape change does not take place (Dawes-Hoang et al., 2005). These

findings strongly indicate that a physical link between AJs and the cytoskeleton must somehow

exist.

INTRODUCTION

9

1.3 Src kinases

Src family kinases (SFKs) are known to be important for several cell biological processes, e.g. cell

migration, cell-shape changes, cell-substratum and cell-cell interactions. SFKs are considered to

function in the modulation of the Actin based cytoskeleton, which represents a determinant of

cell shape change and cell migration (Boschek et al., 1981; Brown and Cooper, 1996).

Furthermore, Src activity is involved in the alteration of the cadherin-catenin complex as tyrosine

phosphorylation of β-Cat or other AJs associated proteins causes weakening of the linkage to the

Actin cytoskeleton (Takeda et al., 1995; Lilien et al., 2002). Phosphorylation of the cadherin-

catenin complex correlates with loss of epithelial character, detachment of cells and gain in

invasiveness (Behrens et al., 1993; Hamaguchi et al., 1993; Lilien et al., 2002). Several proteins are

known to bind Src kinases for being substrates for them. Many of them are associated with the

cytoskeleton and AJs (Thomas and Brugge, 1997). Vertebrate Fes/Fer tyrosine kinases share some

substrates with SFKs, among them p120ctn, β-Cat (Piedra et al., 2003) and Cortactin, which is the

activator of Arp2/3 (Wu and Parsons, 1993; Kim and Wong, 1998).

The vertebrate family of Src non-receptor tyrosine kinases comprehends of 9 members. These are

subdivided into three groups: Src, Yes and Fyn. Each group comprises three members which are

widely expressed in a variety of cells (Thomas and Brugge, 1997). The Drosophila genome encodes

for two Src kinases, Src42A, the closest homolog to vertebrate c-Src (Takahashi et al., 1996), and

Src64B (Simon et al., 1985; Takahashi et al., 1996).

Src non-receptor tyrosine kinases are composed of three main domains: an N-terminal Src

homology 3 domain (SH3), a structural motif known to associate with proline rich regions, a Src

homology 2 domain (SH2) for binding phosphotyrosine, followed by the tyrosine kinase domain.

Other structural features of Src kinases are a myristoylation site at the N-terminus, which is

functioning as a membrane anchor, an autophosphorylation site which is important for activation

and a second tyrosine phosphorylation site at the C-terminus, which is targeted by C-terminal Src

kinase (Csk), an endogenous Src inhibitory factor (Ia et al., 2010). Phosphorylation results in an

intramolecular binding of Src, where the SH2 domain binds to this phosphotyrosine, resulting in a

conformational change which inactivates the kinase (Engen et al., 2008). The domain structure of

Drosophila Src42A is depicted in the results section (see Fig.31 A) and Fig.5 shows the common

structure of Src kinases.

INTRODUCTION

10

Fig.5: Structure of Src kinases

Figure depicts overall domain structure of c-Src kinase, the closest homolog of Drosophila Src42A. The N-

terminus shows a myristoylation site, important for membrane anchoring. An SH3 domain is located in the

N-terminal part, followed by an SH2 domain, important for intramolecular binding of the C-terminal

tyrosine, after Csk dependent phosphorylation. The tyrosine kinase domain locates at the C-terminus of the

kinase. Identfied mutations in v-Src are indicated. Adapted from Parsons and Parsons, 2004.

Members of SFKs are good candidate genes for the regulation of AJ remodelling. In cultured

epithelial cells activated Src was shown to downregulate E-Cad, thereby leading to dissociation of

cells, a process also referred to as epithelial mesenchymal transition (EMT) (Behrens et al., 1993;

Boyer et al., 1997; Thomas and Brugge, 1997).

Drosophila Src42A localizes along the plasma membrane in epithelial cells, whereas activated

Src42A (pSrc) colocalizes with DE-Cad/Arm at the AJs. Evidence was provided by Shindo et al., that

Src42A is preferentially activated at AJs of epithelia undergoing morphogenetic rearrangements.

They showed that Src42A can influence DE-Cad in two distinct, and disparate, ways. First it

antagonizes DE-Cad mediated cell adhesion, while on the other hand positively influencing the

transcription of DE-Cad in a TCF dependent manner. These findings propose a model where

activation of Src42A at the AJs is mediating AJs turnover, thereby promoting their rearrangement

and remodelling of the epithelial tissue. With regard to this it was shown that expression of

activated Src42A increased expression of Escargot (Esg), which is a target of Wg/Arm signaling in

the trachea, whereas mutants for Src42A showed reduced Esg expression. This suggests that

Src42A is acting through the Arm/TCF pathway, because this phenotype was suppressed by co-

expression of dominant negative TCF (TCFΔN) (Chihara and Hayashi, 2000; Llimargas, 2000;

Shindo et al., 2008). However, the function of Srcs in Wg signaling appears to be limited, since

double mutants for Src42A and Src64B do not exhibit segmentation defects, which is a

characteristic of mutations in these genes.

INTRODUCTION

11

Src42A and Src64B were shown to have redundant functions in germband retraction and dorsal

closure (Tateno et al., 2000), which is a process where two lateral epithelial cell sheets migrate

towards each other closing the big gap at the dorsal side which remains after germband retraction

(see further down for more details). Double mutants frequently exhibit broken longitudinal tracts

and commissures, and optic lobe/Bolwig’s organ and trachea formation was found to be

disrupted as well (Takahashi et al., 2005). In comparison, the respective single mutants do not

exhibit severe defects in these processes/structures. In this context, Src42A and Src64B have been

shown to interact genetically and functionally with shotgun, which encodes for DE-Cad, and arm.

Here Src42A and Src64B can trigger cytosolic and nuclear accumulation of Arm. Co-IP experiments

revealed that DE-Cad and Arm form a ternary complex with Src42A (Takahashi et al., 2005; Shindo

et al., 2008). Upon Src42A knockdown it was shown that Arm remained at cellular junctions,

whereas nuclear as well as cytosolic fractions were lower in comparison to the wt situation

(Desprat et al., 2008). Src42A and Src64B functions had been shown to play roles in

WNT5/Derailed signaling, as double mutants for Src42A and Src64B exhibit comparable

commissural phenotypes similar to Wnt5 and derailed mutants (see also above), which could be

suppressed or enhanced by Src gain- and loss-of-function, respectively. A physical interaction

between Derailed and Src64B had been shown in this context as well (Wouda et al., 2008).

As mentioned above, Src42A and Src64B have been shown to have some redundant functions

with regard to morphogenetic processes like dorsal closure. However, some functions have been

shown, where only one single Src kinase is involved. For example mutations in Src64B result in

reduction in female fertility, which is due to nurse cell fusion and ring canal defects (Dodson et al.,

1998), whereas Src42A is supposed to have just minor, if at all, functions during oogenesis

(Takahashi et al., 2005). Src64B was also shown to be important for proper cellularization of the

Drosophila embryo (Thomas and Wieschaus, 2004; Strong and Thomas, 2011). In contrast Src42A

was confirmed to modulate mitochondrial Citrate synthase (CS) activity negatively in vivo, as

mutants show increased CS activity (Chen et al., 2008). Src42A mutants show high frequency of

lethality before hatching, whereas Src64B single mutants are viable (Dodson et al., 1998; Lu and

Li, 1999; Tateno et al., 2000; Takahashi et al., 2005; O’Reilly et al., 2006). However one

hypomorphic Src42A allele is reported (Src42AJP45) which shows some escapers exhibiting mild

dorsal cleft phenotypes (Tateno et al., 2000). Src42A was shown to regulate receptor tyrosine

kinase (RTK) signaling and JUN Kinase (JNK) activity (Lu and Li, 1999; Tateno et al., 2000). Src42A

single mutants exhibit defects in mouthpart formation (Tateno et al., 2000) and defects in leading

INTRODUCTION

12

edge cells: the actomyosin cable is disrupted, phosphotyrosine levels are weaker and dorsal

closure is slightly defective, where 8% show small holes at embryonic stage 16, where the dorsal

hole should be already closed (see Fig.7 B) (Murray et al., 2006). Transcripts of Src42A accumulate

in high levels in neighboring cells upon wound induction and wound-induced genes like Ddc and

ple show widespread wounding induced transcription in Src42A mutants (Juarez et al., 2011). It

was shown that Src42A is acting cell autonomously and inhibiting Ddc expression when its

constitutively active form is expressed.

Src42A was recently shown by two groups to be involved in the elongation of the dorsal trunk of

the tracheal network (Förster and Luschnig, 2012; Nelson et al., 2012; Ochoa-Espinosa et al.,

2012). Src42A single mutants, as well as expression of dominant negative Src42A (Src42AKM),

leads to a shortened dorsal trunk. Expression of Src42A, as well as its constitutively active form,

leads to an extended dorsal trunk respectively. DE-Cad recycling at AJs is affected in Src42A single

mutants, indicating that defective junction remodelling leads to cell shape changes. The apical

surface area of Src42A mutants is significantly reduced. Src42A dependent anisotropic expansion

along the longitudinal axis was shown to be a main driving force for elongation and overall apical

expansion. Furthermore, it was demonstrated that this expansion process is cell autonomous by

expressing Src42A transgenes via the UAS/Gal4 system in three different compartments in the

Src42A mutant background. Expansion had been shown consequently in expressing cells (Förster

and Luschnig, 2012). The short trunk phenotype of Src42A single mutants is epistatic to several

genes which are involved in dorsal trunk development, and overelongated dorsal trunk

phenotypes of respective mutants is not due to increased Src42A activity, indicating a parallel or

downstream pathway where Src42A acts. Fig. 6 depicts the model of these new findings.

INTRODUCTION

13

Fig. 6: Model of dorsal trunk elongation with regard to Src42A function

(A) Shown in blue is the tracheal dorsal trunk. After stage 14 of embryogenesis the dorsal trunk elongates,

which is depicted in the wt embryo. Mutants for Src42A show a shortened dorsal trunk phenotype. The

magnified area indicates the function of Src42A in the anisotropic expansion of dorsal trunk cells in the

longitudinal axis. (B) Model summarizes findings of Förster and Luschnig, 2012 and Nelson et al. of how

Src42A acts in apical membrane growth, as well as in the cell shape changes. Adapted from Ochoa-Espinosa

et al., 2012.

The dorsal closure defects, which have been observed in double mutants for Src42A and Src64B,

indicate functional redundancy with regard to this morphogenetic process. Dorsal closure is the

last big morphogenetic process during Drosophila embryogenesis, where two epidermal lateral

sheets extend to the dorsal side meet and fuse, thereby closing the big dorsal hole which remains

after germband retraction. During dorsal closure the amnioserosa and yolk sac are enclosed inside

the embryo as a consequence. The process where the leading edge cells meet at the dorsal

midline is regulated in part through the remodelling of adherens junctions, which is leading to

INTRODUCTION

14

their fusion. An important signaling pathway for dorsal closure was shown to be JNK signaling

(Jacinto et al., 2002). Fig.7 depicts the process of dorsal closure.

Src42A is proposed to act upstream of JNK signaling. Members of the SFK family cooperate to

regulate JNK activity: double mutants in Src42A and tec29 as well as Src42A and Src64B (as

described above) give dorsal open phenotypes, whereas single mutants do not (Tateno et al.,

2000; Takahashi et al., 2005). Furthermore tec29 Src42A double mutants show loss of dpp and puc

expression at the leading edge, which are downstream effectors of JNK signaling (see Fig.7 C)

(Tateno et al., 2000). Mutations of dfer and Src42A together are causing total failure of dorsal

closure (Murray et al., 2006).

Src42A was shown to act together with DCas in integrin-dependent effector pathways.

Simultaneous reduction of Src42A and DCas functions caused blistered wing phenotypes in adult

escapers. This phenotype had been reported for mutants in the integrin subunits multiple

edematous wings (mew) and inflated (if) as well (Bloor and Brown, 1998), and embryonic cuticles

displayed dorsal closure and anterior cuticle defects (Tikhmyanova et al., 2010). Analysis of Src

and Focal adhesion kinase (Fak56) revealed overlapping and distinct contributions in inhibiting

neuromuscular junction growth, which is transduced by the integrin signaling pathway (Tsai et al.,

2008). Src42A was also shown to be important for the Draper pathway. Here association of Shark

and Draper is mediated by Src42A, since Draper is a Src substrate. This binding promotes

activation of downstream phagocytic signaling events (Ziegenfuss et al., 2008).

All these data nicely demonstrate that Src non-receptor tyrosine kinases, as well as SFKs in

general, are implicated in many different cellular and morphogenetic processes, where AJs are

undergoing rearrangements and are remodelled. Many of those genes do not exhibit dramatic

phenotypes as in contrast their combinations do. This demonstrates that these kinases have many

overlapping functions, indicating a highly dynamic and complex network.

INTRODUCTION

15

Fig.7: Dosal closure and JNK signaling

(A) Shown is an embryo at developmental stage 14, where a big dorsal hole remains as a consequence of

germband retraction. Amnioserosa cells (AS) are marked in green, leading edge cells (LE) are labeled in red

which represent the most dorsal epithelial cell row. (B) After embryonic stage 15, the dorsal hole is closed

by the process of dorsal closure, and both leading edge cell rows build a seam at the dorsal midline. (C) JNK

signaling is important for dorsal closure. The result of the JNK pathway is the secretion of Dpp at the leading

edge and expression of puc, which encodes a dual phosphatase dephosphorylating Bsk (JNK) in a negative

feedback loop. Adapted from VanHook and Letsou, 2008.

INTRODUCTION

16

1.4 LMO7

The family of PDZ and LIM domain containing proteins comprises ten members possessing PDZ

domains and at least one LIM domain. LIM domains (name derived from C.elegans lin-11, rat ISL-1

and C.elegans mec-3) (Way and Chalfie, 1988; Freyd et al., 1990; Karlsson et al., 1990) act as

protein-protein binding interfaces. The domain is approximately 55 amino acids in size and

characterized by a highly conserved histidine/cysteine motif important for the binding of two zinc

ions, thereby forming a two-zinc-finger-like structure (Bach, 2000; Kadrmas and Beckerle, 2004;

Te Velthuis et al., 2007). PDZ and LIM family proteins are thought to be involved in Z-Band

formation of muscles through their PDZ domains, which can bind to α-Actinin or β-tropomyosin.

However, PDZ and LIM domain containing proteins are associated with the cytoskeleton directly

or indirectly as well (Harris and Lim, 2001; Kadrmas and Beckerle, 2004). The group of PDZ and

LIM domain encoding genes consists of four subgroups: the ALP subfamily (ALP, Elfin, Mystique,

and RIL), the Enigma subfamily (Enigma, Enigma Homolog, and ZASP), LIM kinases (LIMK1 and

LIMK2), and the LIM only protein 7 (LMO7). The latter protein will be discussed in more detail.

LMO7 was initially linked as a candidate gene to breast cancer progression, due to implication of

human genomic region 13q21-22 in cancer development (Rozenblum et al., 2002) and was found

to be upregulated in several human tumors, among them lymphnode metastasis in breast cancer

(Sasaki et al., 2003).

LMO7 contains an intramolecular PDZ domain, a C-terminal LIM domain and a Calponin homology

domain (CH). A partial consensus sequence for a putative F-box motif has been described earlier

(Cenciarelli et al., 1999), which fails to be detected by current prediction programs (Te Velthuis et

al., 2007). Coiled coil domains are predicted for some LMO7 gene products as well, which is

species dependent. So far, functional analysis of these domains has not been reported for LMO7.

The LMO7 gene (on chromosome 13q22 in humans, see above) was duplicated through evolution

and the gene product of its paralog LIMCH1 (on chromosome 4p13 in humans) shows 64%

identical amino acid sequence of the CH domain and 60% homology of the LIM domain,

respectively. However, LIMCH1 does not contain an additional PDZ domain. Beside these domains

three regions with high homology have been identified within LMO7 and LIMCH1. These regions

may indicate the existence of domains within these proteins, which have not been identified yet

(Friedberg, 2009, 2010). LIMCH1 was found to be upregulated in PIK3CA-mutated tumors (Cizkova

et al., 2010). LIMCH1 was described to be expressed in the presomitic mesoderm, however,

targeted mutations for LIMCH1 have not been reported yet (Sewell et al., 2010). LMO7 exhibits a

INTRODUCTION

17

splice variant lacking the LIM domain in mice brain cDNA (Tanaka-Okamoto et al., 2009), whereas

LIMCH1 shows splice variants lacking the CH domain (Friedberg, 2009). Homologs of LMO7 have

been identified in invertebrates. temporarily assigned gene 204 (tag204) was found to be the

LMO7 homolog in C.elegans and CG31534 (now annotated as CG43427) encoding the Drosophila

homolog respectively. These homologs show high conservation in regard to the C-terminal LIM

domain but the invertebrate homologs do not exhibit PDZ and CH domains (Te Velthuis et al.,

2007). A domain structure of full length vertebrate LMO7 is shown in Fig.9 A.

Studies indicated that LMO7 is likely involved in the formation and maintenance of epithelial

architecture by remodelling the Actin cytoskeleton. The LIM domain has been shown to interact

with Afadin (which in turn associates with Nectins). LMO7 binds to α-Actinin (an Actin binding

protein). These interactions are thought to modulate a link between the cell adhesion complex of

E-Cad and the Nectin network. Furthermore Afadin can directly associate with F-Actin, therefore

creating a second link between LMO7 and the Actin cytoskeleton. Antibodies against LMO7

showed expression in various rat tissues including the heart, lung, small intestine, kidney, brain,

liver, spleen, and skeletal muscle. Staining of LMO7 showed colocalization with Afadin in the

region of the AJs in epithelial cells of rat gallbladder (see Fig.8), supporting the biochemical data.

It was furthermore shown that E-Cad, β-Cat and α-Cat co-immunoprecipitate with LMO7, even in

afadin-/- ES cells, supporting the hypothesis that LMO7 connects Nectins with the E-Cad adhesion

complex. Whether LMO7 can directly associate with the Actin cytoskeleton remains unclear, but

CH domains can bind to Actin bundles directly, thereby suggesting a role of LMO7 in direct

association with the Actin cytoskeleton (Ooshio et al., 2004).

INTRODUCTION

18

Fig.8: Localization of LMO7 in epithelial cells of rat gallbladder

(A) Localization of LMO7 in epithelial cells of rat gallbladder. Afadin localizes to AJs, where LMO7 is detected

as well (arrows), slightly basal to the TJ marker ZO-1. LMO7 is additionally detected at the cytoplasmic faces

of the apical membrane (arrowheads). Scalebar represents 10 µm. (B) Immunoelectron microscopy

revealed that LMO7 localizes to the AJs (arrows) and at the cytoplasmic faces of the apical membrane

(arrowheads). Scalebar represents 0.1 µm. Adapted from Ooshio et al., 2004.

A large deletion of around 800 kb, covering Uchl3 and LMO7 gene loci, resulted in lethality for

about 40% of mice between birth and weaning and surviving homozygotes showed muscular

degeneration and growth retardation, as well as retinal degeneration. The latter phenotype is

suggested to be caused likely by the Uchl3 knockout. Respective single mutants show the same

retinal degeneration defects. Although the proportion of muscles to body weight was

proportionate in the Uchl3 LMO7 knockout mice, nuclei were elongated and increased in their

INTRODUCTION

19

amount by a factor of 2 in thigh muscle fibers (Semenova et al., 2003). These observations are

consistent with an expression analysis performed in mice, where LMO7 mRNA was detected in

somites and the eye, respectively (Ott et al., 2008).

LMO7 knockdown in the zebrafish Danio rerio causes defects in the cardiac conduction system,

including arrhythmia and heart delocalization. The latter could indicate a possible function of

LMO7 for neural crest cells and their migration. Severe defects which had been observed were

shorter embryos and strongly bent tails. Severe elongation defects were found later in

development. Some extreme phenotypes exhibited upon LMO7 knockdown were head defects.

Rescue experiments showed that these phenotypes were observed upon the morpholino

injection. The knockdown was targeted against the 5’UTR of LMO7, whereas LMO7 RNA was

coinjected without the respective UTR (Ott et al., 2008).

LMO7 is transcribed in the lung, heart, brain and kidney. An alternative splice form, lacking the

LIM domain, was identified in a mouse brain cDNA library (Tanaka-Okamoto et al., 2009). It was

shown, that transforming growthfactor-β1 (TGF-β1) induces expression of LMO7 while enhancing

invasiveness of rat ascites hepatoma cells. Furthermore, TGF-β1 induced this alternatively spliced

variant of LMO7S, lacking the C-terminal LIM domain (Nakamura et al., 2005). LMO7 localizes to

the luminal surface of epithelial cells. The PDZ domain is essential for the apical localization,

because LMO7 deficient mice lacking the PDZ domain showed cytosolic mislocalization of LMO7.

These LMO7 knockout mice were viable and fertile, and had been indistinguishable in

appearance, size, growth, development and behaviour from their littermates. Lung sections of 14-

week old LMO7 deficient mice showed irregular epithelial sheets, respiratory bronchioles and

alveolar ducts. However, although the position of AJs was slightly deviated, E-Cad and its

associated Catenins, as well as Afadin and Nectins were localizing at the AJs, indicating that LMO7

function is not required for proper AJs formation. Mice at an age of 90 weeks deficient for LMO7

showed development of adenocarcinomas to an extent of about 22%, whereas LMO7+/- mice

developed lung cancer to 13%. It was shown that cultured tumor cell lines deficient for LMO7

possess chromosome abnormalities and cause tumor formation in vivo when injected into nude

mice. These observations indicate tumor suppressor roles for LMO7 (Tanaka-Okamoto et al.,

2009). With regards to these observations, human lung adenocarcinomas showed that LMO7

expression was decreased with tumor progression (Nakamura et al., 2011).

INTRODUCTION

20

Beside the tumor suppressor functions of LMO7 and its localization at the AJs, LMO7 was

furthermore shown to be involved in gene expression. Upon LMO7 knockdown in cell culture it

was found that around 4000 genes showed altered expression. Among these the muscle relevant

genes emerin was identified (Holaska et al., 2006). LMO7 was reported to bind the LEM domain

protein Emerin, in which mutations cause Emery-Dreifuss muscular dystrophy. This disease was

reported as well for mutations in LMNA, which encodes an A-type lamin (Nagano et al., 1996;

Emery, 2000; Bengtsson and Wilson, 2004). LMO7 activates the transcription of muscle-relevant

genes as well as the expression of the emerin gene itself (see above). The binding of LMO7 to

Emerin inhibits LMO7 function in emerin expression, indicating a negative feedback mechanism

(Holaska and Wilson, 2006; Holaska et al., 2006). Furthermore Emerin is required for nuclear

localization of LMO7. Emery-Dreifuss muscular dystrophy has been linked to LMO7 since one

isolated missense mutant of emerin (P183H), is deficient in LMO7 binding. However, three other

gene products of mutations in emerin were still found associating with LMO7 (Holaska et al.,

2006). Recently it was shown that LMO7 can directly bind to the promoters of Pax3, MyoD and

Myf5, suggesting that LMO7 is directly involved in their expression. This interaction is suppressed

by Emerin, providing a mechanism of how Emerin inhibits LMO7 function (Dedeic et al., 2011).

Fig.9 B shows a summarized model of LMO7 functions in cell-cell adhesion as well as its nuclear

role in expression of muscle-relevant genes.

INTRODUCTION

21

Fig.9: LMO7 structure and function

(A) Structure of full length LMO7. CH domains were identified at the N-terminus of the protein, a PDZ

domain is found as well as a C-terminal LIM domain. Described splice variants of LMO7 are not shown. (B)

Summarized functions of LMO7 at AJs and in the nucleus. LMO7 associates with α-Actinin at one hand while

binding to Afadin at the other. Since Afadin associates with F-Actin and α-Actinin with Catenins, LMO7 is

suggested to build a link between the E-Cad adhesion complex and Nectins, which bind to Afadin. LMO7

might be able to associate directly with Actin bundles as well by the CH domain. A second function of LMO7

is its involvement in the expression of muscle-relevant genes and emerin. The gene product of emerin was

furthermore shown to bind LMO7, thereby inhibiting expression of target genes. This is most likely by

recruiting LMO7 to the nuclear envelope, where Emerin localizes. LMO7 was shown recently to directly bind

promoters of Pax3, MyoD and Myf5. Adapted from Te Velthuis and Bagowski, 2007.

1.5 smallish (CG43427)

As already described in the first two chapters of the introduction, baz gene function was found to

be implicated in the establishment of cell polarity of many types of tissues (e.g. ectodermal

epithelia and NBs) as well into the formation of AJs of epithelial cells. Baz was found in a complex

with DaPKC and DPar6, referred as the Par complex (Wodarz et al., 2000; Petronczki and Knoblich,

2001). The Par complex is important for the apical membrane identity at first, whereas release of

Baz by DaPKC phosphorylation causes relocalization of Baz at the AJs, where Baz was shown to

recruit cadherin-catenin clusters, thereby being implicated in the formation of these cell-cell

contact sites (Wei et al., 2005; McGill et al., 2009). As baz gene function was shown to be epistatic

to other polarity markers it is suggested that Baz is a key player in mediating cell polarity. As the

past years revealed several new binding partners of Baz, thereby identifying new cellular

pathways were Baz is functioning, its overall role remains elusive in many aspects.

To unriddle the function of Baz in the establishment of cell polarity in more detail, a yeast two-

hybrid screen was performed for the identification of new binding partners of the protein

(Ramrath, 2002). Three different regions of Baz were chosen as baits, among them the N-terminal

oligomerization domain (Benton and Johnston, 2003), the C-terminus, as well as the

intramolecular region encoding for the three PDZ domains. The latter one is of interest for this

work, because one potential binding partner identified was binding to this PDZ domain containing

region within Baz. The potential interactor identified as prey was the C-terminus of the annotated

INTRODUCTION

22

gene CG31534, which was not described so far. The C-terminus encoded for a LIM domain and a

PDZ binding motif of class I (S/T X‡Φ§ -COOH), which made this hit very interesting. PDZ binding

motifs are the counterpart of PDZ domains, which can bind into a hydrophobic pocket within the

PDZ domain (Harris and Lim, 2001). Furthermore CG31534 turned out to be likely the Drosophila

homolog of vertebrate LMO7 (see 1.4), which was already implicated functioning at AJs in

vertebrates. A detailed scheme of the yeast two-hybrid screen is shown in the results section (see

Fig.12).

CG31534 is located on the right arm of the third chromosome. However, in 2008 only two splice

variants were annotated for CG31534: one encoding for 889aa (98.8 kDa), which resembled the

full length form of the protein and a second isoform encoding for 849aa (94.4 kDa) respectively.

This isoform was lacking the LIM domain by use of an alternative spliced exon containing a

premature stop signal, which was also reported for a LMO7 (LMO7S) specific neuronal isoform

(Nakamura et al., 2005). By 2009 the gene prediction for CG31534 was changed in a way that four

isoforms were annotated, encoding for two additional isoforms. They were slightly deviated from

the two isoforms mentioned above with minor changes affecting the C-terminus, thereby

exhibiting 8 additional amino acids in the region N-terminal to the LIM domain, as well as a

change affecting the N-terminus of the full length isoform by use of an alternative ATG, causing a

slightly larger protein. Altogether CG31534 was encoding four protein isoforms, CG31534-PA,

CG31534-PB, CG31534-PC (857aa, 95.3 kDa) and CG31534-PD (932aa, 103.5 kDa), respectively. Of

major interest had been only the two protein isoforms PA and PD due to the lack of the C-

terminal region carrying the LIM domain and the PDZ binding motif in the two isoforms PB and

PC, which was identified in the yeast two-hybrid screen as prey.

Unluckily, the gene annotation of CG31534 was still not correct at this time. In 2011 (nearly at the

end of the regular time limit of this work given by the GGNB doctoral program), a new gene

annotation release of flybase (gene annotation release 5.40) indicated that the neighboring gene

CG31531, located 5’ to CG31534, is part of the same transcription unit. We confirmed by PCR,

using embryonic cDNA as template, that both transcription units indeed resemble a single gene,

spanning approximately 52 kb (3R 485,301 – 537,915). A detailed gene map of the current gene

annotation of CG43427 is shown in Fig.17 B in the results section. Most biochemical data

produced within this work were performed with the shorter protein isoform CG31534-PA, which

is now CG43427-PI, reflecting only two third of the C-terminus of the full length protein. More

recent data generated were performed using the larger isoform, annotated as CG43427-PM,

INTRODUCTION

23

encoding for 1533aa (168.9 kDa) and additionally possessing two coiled coil domains in the N-

terminal region (see also Fig.12 in the results section).

At the start of the thesis some preliminary data had been obtained in two diploma works before

(Neugebauer, 2007; Beati, 2009). Endogenous CG31534 protein was detected apical in

ectodermal epithelia, which was already a good hint. In cell culture experiments it was shown that

Baz can recruit CG31534, which localized in the cytoplasm without Baz, to the cell cortex. In vivo

binding of Baz and CG31534 was shown later by Co-IP experiments using embryonic lysates

expressing an N-terminally GFP tagged version of CG31534-PA. However, mutations generated in

both works for CG31534 were viable and fertile (Beati, 2009), which is also a characteristic for the

respective LMO7 knockout in mice (Tanaka-Okamoto et al., 2009).

SCOPE OF THE THESIS

24

1.6 Scope of the thesis

The aim of the thesis was the analysis of the gene function of CG43427 (smallish, (smash)) during

the development of Drosophila. Smash was identified as a new potential binding partner of the

cell polarity regulator Baz (Ramrath, 2002). Preliminary work showed that endogenous Smash

protein colocalizes with Baz in the region of the AJs. As both proteins associate in vivo, we wanted

to further analyse the function of smash with regard to epithelial cell polarity. A knockout of

smash was generated before with regard to the old gene annotation. This allele is viable and

fertile but did not show polarity defects. This allele cannot be considered as a classical null,

because N-terminal parts might be still expressed. Thus we generated a mutation for smash,

affecting the entire genomic locus by FLP/FRT mediated transdeletion. This full knockout allele

was analysed for its viability and for potential polarity defects. Flies lacking the entire genomic

locus are still viable and fertile and do not show obvious defects.

All preliminary data had been obtained with the short isoform Smash-PI. As smash encodes a

larger isoform we wanted to analyse potential gain of function phenotypes. Accordingly, the

respective smash isoform had to be cloned and transgenic flies were subsequently generated.

Overexpression of the respective large isoform caused a dramatic increase in the lethality score

and embryonic cuticles showed anterior and dorsal holes.

Beside the interaction of Smash and Baz, other binding partners of Smash had been of interest.

Preliminary data showed that Smash binds to the non-receptor tyrosine kinase Src42A (Beati,

2009). Based on this finding, we wanted to continue to investigate the developmental relevance

of this interaction. Src42A has been implicated to function in dorsal closure and other

morphogenetic processes. Thus we analyzed whether smash might also function in pathways

coordinating dorsal closure. With regard to this we focussed on Src64B as well. Src42A is known to

function redundantly with Src64B in morphogenetic processes like dorsal closure (see 1.3). Of

interest had been double mutant combinations of smash with Src42A or Src64B respectively. A

double mutant with Src64B is lethal. However, cell polarity was not affected and embryonic

cuticles did not show the dorsal open phenotype.

MATERIAL AND METHODS

25

2 Material and methods

2.1 Chemicals and materials

2.1.1 Chemicals and enzymes

Used chemicals were purchased from following companies:

Acros, Geel, Belgium; Baker, Deventer, Netherlands; Biomol, Hamburg, Germany; Bio-RAD,

Munich, Germany; Difco, Detroit, U.S.A.; Fluka, Buchs, Swiss; Gibco/BRL Life Technologies;

Karlsruhe, Germany; Gruessing, Filsum, Germany; Merck, Darmstadt, Germany; Riedel-de Haên,

Seelze, Germany; Roth, Karlsruhe, Germany; Serva, Germany; Sigma-Aldrich, Steinheim, Machery-

Nagel, Dueren, Germany.

Demineralized water was used for solutions, buffers, etc., which were autoclaved if necessary.

Enzymes were purchased from following companies:

Boehringer/Roche Diagnostics, Mannheim, Germany; MBI Fermentas, St. Leon-Rot, Germany;

New England Biolabs, Schwalbach-Taunus, Germany, Invitrogen, Karlsruhe, Germany; Promega,

Madison, USA.

2.1.2 Kit systems

The following kits were used in this work:

Nucleobond AX100, Macherey-Nagel

NucleoSpin Extract II, Macherey Nagel

pENTRTM/D-TOPO® Cloning Kit, Invitrogen

Gateway® LR ClonaseTM II Enzyme Mix, Invitrogen

BM Chemiluminescence Blotting Substrate, Roche Diagnostics


26

2.1.3 Photo and picture analysis

Light microscopy: Axio Imager, Carl Zeiss Jena GmbH, Germany

Fluorescence binocular: Leica MZ 16 FA

Confocal microscopy: LSM 510 Meta, Carl Zeiss Jena GmbH, Germany

X-ray films: Fuji Medical X-Ray Film „Super RX“, Fuji, Tokyo, Japan

X-ray film developer and fixer: Tenetal Roentogen, Tenetal, Norderstedt, Germany

Operating Systems: Macintosh iMac, Apple, Ismaning, USA

Microsoft Windows XP and Windows Vista, Microsoft,

Redmont, USA

Image processing: GIMP, GNU General Public License (GPL)

Inkscape, GNUGeneral Public License (GPL)

IrfanView (Proprietary Freeware)

Sequence und primer analysis: DNA-Star Lasergene V7, DNASTAR Inc.Madison, USA

2.1.4 Bacterial strains and cell culture lines

Bacterial strains and cell culture lines which were used for this work are listed in Table 1.

Table 1: Used bacterial strains and cell lines

Cell line Usage

E.coli DH5α Amplification and purification of plasmid DNA

E.coli XL-1 blue

E.coli Bl21 Expression and purification of recombinant proteins

E.coli Top 10 one shot Transformation of DNA after pENTRTM

/D-TOPO® cloning reaction

S2 cells Transfection for biochemical experiments


27

2.1.5 Plasmids

Used plasmids and their properties are listed in Table 2.

Table 2: Used plasmids and their properties

Plasmid Property

pENTRTM/D-TOPO®

plasmid for gateway cloning; insertion of PCR products with 5' CACC overhang

pPGW Destinationvector for gateway cloning; N-terminal tagged with GFP; under control of UASp promotor

pPWG Destinationvector for gateway cloning; C-terminal tagged with GFP; under control of UASp promotor

pUASpGW Destinationvector for gateway cloning; N-terminal tagged with GFP; under control of UASp promotor; contains attp

site

pUASpWG Destinationvector for gateway cloning; C-terminal tagged with GFP; under control of UASp promotor; contains attp

site

pPWH Destinationvector for gateway cloning; C-terminal tagged with HA; under control of UASp promotor

pHGWA Destinationvector for gateway cloning; N-terminal tagged with His and GST + C-terminal His tag; under control of

T7 promotor

2.1.6 Buffers and medium

Buffers and reagents for histology:

PBS: 140 mM NaCl

10 mM KCl

2 mM KH2PO4

6.4 mM Na2HPO4 x 2H2O

pH 7.3

4% Formaldehyde in PBS: 39.75 ml PBS

5.25 ml 37% Formaldehyde

1x Triton salt solution: 1l dH2O

0.3 ml Triton X-100

4 g NaCl


28

PBTw: PBS with 0.1% Tween20

NHS: Normal Horse Serum (Gibco)

Mowiol: 5 g Mowiol

20 ml PBS

10 ml Glycerol

Hoyers mountant: 30 g Gumarabic

50 ml dH2O

200 g Chloralhydrate

20 g glycerol

Before use 800 µl hoyers mountant were gently mixed with 640 µl lactic acid.

Buffers for molecular biological methods:

TE-Buffer: 10 mM Tris HCl

0.5 mM EDTA

pH 8.0

TAE-Puffer: 40 mM Tris acetate

Acetic Acid

1 mM EDTA

pH 7.4

Buffers for DNA purification from E.coli:

S1 buffer: 50 mM TrisHCl

10 mM EDTA

100 μg/ml RNase A

pH 8.0


29

S2 buffer: 200 mM NaOH

1% SDS

S3 buffer: 2.8 M Potassium acetate (CH3CO2K)

pH 5.1

Buffers S1 and S3 are stored at 4°C.

Buffers for DNA purification from flies:

Squishing buffer: 25 mM NaCl

1 mM EDTA

10 mM TrisHCl

pH 8.0

Quick Fly Genomic DNA Prep

Buffer A: 100 mM TrisHCl

pH 7.5

100 mM EDTA

100 mM NaCl

0.5% SDS

Buffer for protein biochemical experiments:

Lysis buffers

LLBVII buffer: 150 mM NaCl

1% Igepal

50 mM Tris HCl

pH 8.0


30

TNT buffer: 150 mM NaCl

50 mM Tris HCl

1% Triton X-100

pH 8.0

TNT buffer (high salt): 500 mM NaCl

50 mM Tris HCl

1% Triton X-100

pH 8.0

Proteinase inhibitors:

1 mg/ml Aprotinin

1 mg/ml Leupeptin

1 mg/ml Pepstatin A

0.5 M Pefabloc SC

Proteinase inhibitors were added to the lysis buffer in a concentration of 1:500.

Phosphatase inhibitors:

500 mM Sodiumorthovanadate

50 mM Phenyloxide

Halt Phosphatase Inhibitor Cocktail (Thermo Scientific)


31

Phosphatase inhibitors were added to the lysis buffer to a concentration of 1:500, phosphatase

inhibitor cocktail was used 1:100.

SDS gelelectrophoresis and Western blotting:

2x SDS loading buffer: 0.2% Bromophenolblue

200 mM beta mercaptoethanol

20% glycerol

4% SDS

100 mM Tris HCl

pH 6.8

1x SDS buffer: 192 mM glycine

25 mM Trisbase

0.1% SDS

Transfer buffer: 25 mM TrisHCl

192 mM glycine

20% (v/v) methanol

TBST: 150 mM NaCl

1 mM Tris HCl

0.2% Tween20

pH 8.0


32

Blocking buffer: TBST containing 3% skim milk and 1% BSA

Blocking buffer: TBST containing 5% BSA

Culture mediums:

LB medium: 0.5% (w/v) yeast extract

1% Trypton

1% NaCl

SOC-Medium: 20 g Trypton

5 g yeast extract

0.5 g NaCl

10 ml 0.25 M KCl

5 ml 2 M MgCl2

20 ml 1 M glucose

filled up to 1l volume with dH2O

pH 7.0

S2 medium: Schneider's Drosophila Medium (Gibco)

10% Serum, insect cell tested (Sigma)

PenStrep (Gibco)

2.1.7 Primers

Table 3 lists primers that were used for this work. Primers were ordered from metabion

international AG, Martinsried and BioTez GmbH, Berlin.


33

Table 3: Used primers and their application

name sequence [5' - 3'] [°C] application

M13 forward GTAAAACGACGGCCAG 51 sequencing of pENTRTM/D-TOPO®

M13 reverse CAGGAAACAGCTATGAC 47 sequencing of pENTRTM/D-TOPO®

UASp forward GGCAAGGGTCGAGTCGATAG 58 sequencing of gateway® destination vectors

pPTagW rev GGTATAATGTTATCAAGCTC 46 sequencing of gateway® destination vectors

eGFP N rev CGGACACGCTGAACTTGTG 57 sequencing of gateway® destination vectors

eGFP C for CAAAGACCCCAACGAGAAG 54 sequencing of gateway® destination vectors

piggyBac 5' outside CAATTTTACGCAGAC

TATCTTTCTAGGG 57 screening for mutants after transdeletion

CG31534 genomic-e03181-R CCGATATATGCTCACCTCATGC 56 screening for mutants after transdeletion

piggyBac 3' outside CGTACGTCACAATATGATTATCTTTCT

AGG 56 screening for mutants after transdeletion

CG43427 PBACRB test GTCGCGGCGCTGCCATGATCATCG 68 screening for mutants after transdeletion

UP Primer GACGGGACCACCTTATTTTATTCTATC

ATG 59 screening for mutants after transdeletion

CG43427 genomic for P(XP) test

CACACTCCGCCCCATTTTTATATCC 60 screening for mutants after transdeletion

check ATG I CG43427 for GTGCAGATCCATCACGGATGC 60 test for successful deletion of the first ATG of CG43427

check ATG I CG43427 rev GCGCAATGGAAAACTCACCA 58 test for successful deletion of the first ATG of CG43427

check ATG II CG43427 for CATGTCACGCCCACCTCATC 59 test for successful deletion of the second ATG of

CG43427

check ATG II CG43427 rev GCACCGGCCTTAAATGCTTG 58 test for successful deletion of the second ATG of

CG43427

Control_CG9769_ forward

CCGCACTATCGATAGGCCC 58 control primers for downstream neighboring gene

CG9769

Control_CG9769_ reverse

CATGCAGCCGCTCTTGC 58 control primers for downstream neighboring gene

CG9769

Src42A-DeltaSH3-F GCAGGTGCCAACGTC

GGA TCC GGT GGA GGT AAATCAATCGAAGCA

75 deletes SH3 domain, introduces BamHI restriction site

Src42A-DeltaSH3-R TGCTTCGATTGATTTACCTCCA

CCGGATCCGACGTTGGCACCTGC 75 deletes SH3 domain, introduces BamHI restriction site

Src42A-DeltaSH2-F AAATCAATCGAAGCA

GGA TCC GGT GGA GGT GTCCAGATCGAGAAG

72 deletes SH2 domain

Src42A-DeltaSH2-R CTTCTCGATCTGGACACCTCCA

CCGGATCCTGCTTCGATTGATTT 72 deletes SH2 domain

Src42A-DeltaKinase-F ATCGACAGAACATCC GGA

TCC GGT GGA GGT GAAGACTTCTATACA

71 deletes tyrosine kinase domain

Src42A-DeltaKinase-R TGTATAGAAGTCTTCACCTCCA

CCGGATCCGGATGTTCTGTCGAT 71 deletes tyrosine kinase domain


34

Src42A-YF-F ACA TCT GAT CAG AGC GAC

TTC AAA GAG GCG CAA GCC TAC 71

mutates last tyrosine to phenylalanine, deletes StuI restriction site

Src42A-YF-R GTAGGCTTGCGCCTCTTTG

AAGTCGCTCTGATCAGATGT 79

mutates last tyrosine to phenylalanine, deletes StuI restriction site

Mut Y64F for LIM GGACGAGAGCCCATC

TTT GAGAATGTCAGCTCG 68 mutates tyrosine 64 of CG43427-RI to phenylalanine

Mut Y64F rev LIM CGAGCTGACATTCTC

AAA GATGGGCTCTCGTCC 68

Mut Y152F for LIM CAAGCAGAGCCCTAC

TTC CAAGTGCCGAAGGCC 72 mutates tyrosine 152 of CG43427-RI to phenylalanine

Mut Y152F rev LIM GGCCTTCGGCACTTG

GAA GTAGGGCTCTGCTTG 72

Mut Y162F for LIM GCCACGGAGCCCTAC TTC

GATGCCCCCAAGCAT 74 mutates tyrosine 162 of CG43427-RI to phenylalanine

Mut Y162F rev ATGCTTGGGGGCATC GAA

GTAGGGCTCCGTGGC 74

Mut Y244F for LIM TCATCGGACAACACC TTC

GAGACCATATCGAAC 66 mutates tyrosine 244 of CG43427-RI to phenylalanine

Mut Y244F rev GTTCGATATGGTCTC GAA

GGTGTTGTCCGATGA 66

Mut Y601F for LIM GGCCTGGAGCCAGAT TTC

GCTGTTAGCACGAGG 71 mutates tyrosine 601 of CG43427-RI to phenylalanine

Mut Y601F rev LIM CCTCGTGCTAACAGC GAA

ATCTGGCTCCAGGCC 71

Y685F Mut for LIM CAGCAGCGGAAGAGT TTT

GACAGCCAACAGACC 69 mutates tyrosine 685 of CG43427-RI to phenylalanine

Y685F Mut rev LIM GGTCTGTTGGCTGTC AAA

ACTCTTCCGCTGCTG 69

LIM delta prr1 for

GTAGGGCTGATGGAG GCA GCA AAGGAAAAG

GCA GCA GCA GCA GCA ACCGAGAGTCCGATT

79

exchanges prolines in proline rich region 1 to glycines

LIM delta prr1 rev AATCGGACTCTCGGTTGCTGCTG CTGCTGCCTTTTCCTTTGCTGCC

TCCATCAGCCCTAC 79

LIM delta prr2 for CAGTCAGAGGCCACA

GCA GCA GCA TTG GCA GCA GCA GCATCGACCGCCCAAGTG

81

exchanges prolines in proline rich region 2 to glycines

LIM delta prr2 rev CACTTGGGCGGTCGATG

CTGCTGCTGCCAATGCTGCTGCTGT GGCCTCTGACTG

81

Lim seq1 CTACCAAGTGCCGAAGGCCACG 64 sequencing of CG43427-RI

Lim seq2 CTCGCCAGAGCAGTGAGCACTAC 63 sequencing of CG43427-RI

Lim seq3 GCATGCATATCACCAGATGGAC 57 sequencing of CG43427-RI


35

2.1.8 Primary antibodies

Primary antibodies, which were used for immunofluorescence staining (IF), Western blotting (WB)

and Immunoprecipitation (IP), are listed in Table 4. Secondary antibodies and fluorochrome

conjugated phalloidin for F-actin staining are listed in Table 5.

Table 4: Primary antibodies

antibody species application source / company

Actin rabbit WB 1:2000 Sigma A2066

Smash intra (DE02088)

rabbit IF 1:500; WB 1:2000; IP Wodarz, unpublished

Smash N-term (DE12095)

guinea pig

IF 1:500; WB 1:1000 Beati, unpublished

Bazooka (DE99646) rabbit IF 1:2000; WB 1:2000;

IP Wodarz et al., 1999

DE-Cadherin (DCAD2) rat IF 1:5; WB 1:5, IP

Developmental Studies Hybridoma Bank (DSHB)

Discs Large (4F3) mouse IF 1:20

Armadillo (N2 7A1) mouse IF 1:20; WB 1:50

α-Catenin (D-CAT1) rat IF 1:50

GFP mouse IF 1:2000; WB 1:2000;

IP Invitrogen Molecular Probes # A11120

GFP rabbit IF 1:2000; WB 1:2000;

IP Invitrogen Molecular Probes # A11122

HA mouse WB 1:2000 Roche # 11 583 816 001

Phosphotyrosine (PT-66)

mouse IF 1:1000; WB 1:1000 Sigma P3300

Src42A rabbit IF 1:1000 Takahashi et al., 2005

pSrc rabbit IF 1:1000 Shindo et al., 2008

BrdU (MoBu-1) mouse IF 1:500 Abcam (ab8039)

mCherry (1C51) mouse IF 1:500 Abcam (ab125096)

Table 5: antibodies and fluorochrome conjugated phalloidin

antibody species application company

Peroxidase-AffiniPure Goat Anti-Mouse IgG + IgM (H+L)

goat 1:10.000 Jackson ImmunoResearch

(115-035-068)

Peroxidase-AffiniPure Goat Anti-Rabbit IgG, (H+L)


(111-035-144)

Peroxidase-AffiniPure Goat Anti-Rat IgG + IgM, (H+L)


(112-035-068)

Alexa Fluor® 488 Goat Anti-Mouse IgG1 (γ1) goat 1:200 Invitrogen Life Technologies

A21121

Alexa Fluor® 555 Anti-Mouse IgG (H+L) goat 1:200 Invitrogen Life Technologies

A21422


36

Alexa Fluor® 647 Goat Anti-Mouse IgG (H+L) goat 1:200 Invitrogen Life Technologies

A21235

Alexa Fluor® 488 Goat Anti-Rabbit IgG (H+L) goat 1:200 Invitrogen Life Technologies

A11008


A21428


A21244

Alexa Fluor® 488 Goat Anti-Rat IgG (H+L) goat 1:200 Invitrogen Life Technologies

A11006


A21434


A21247

Alexa Fluor® 488 Goat Anti-Guinea Pig IgG (H+L), highly cross-adsorbed

goat 1:200 Invitrogen Life Technologies

A11073



A21435



A21450

Alexa Fluor® phalloidin 488 1:100 Invitrogen Life Technologies

A12379


A12380


A22287

2.1.9 Fly stocks

Used fly stocks and donors are listed below in Table 6.

Table 6: Fly stocks

Stock Plain text genotype Description Reference

Oregon R wildtype Wodarz stock

collection

white1118 w[1118] white eyes Bloomington

#5905

daughterless Gal4

w[1118]; P{da-GAL4.w[-]}3 Gal4 driver line; ubiquitous expression under

control of daughterless promotor; 3rd chromosome

Bloomington #8641


37

engrailed Gal4 y[1] w[*]; P{w[+mW.hs]=en2.4-GAL4}e16E

P{w[+mC]=UAS-FLP1.D}JD1 Gal4 driver line; expression under control of

engrailed promotor in stripes; 2nd chromosome Bloomington

#6356

daughterless Gal4

w[*]; P{w[+mW.hs]=GAL4-da.G32}UH1 Gal4 driver line; ubiquitous expression under

control of daughterless promotor; 2nd chromosome

Wodarz stock collection

tubulin Gal4 Y[1] w[*]; P{w[+mC]=tubP-GAL4}LL7/TM3,

Sb[1] Gal4 driver line; ubiquitous expression under control of tubulin promotor; 3rd chromosome

Bloomington #5138

patched Gal4 W[*]; P{w[+mW.hs]=GawB}ptc[559.1] Gal4 driver line; expression under control of

patched promotor in stripes; 2nd chromosome Bloomington

#2017

hsFlp;; actin > CD2 > Gal4, UAS

RFP

P{ry[+t7.2]=hsFLP}12, y[1] w[*]; P{w[.mC]=Act5C(CD2)Gal4}17bFO1, UAS RFP

/TM3, Sb[1]

Gal4 driver line, expression under control of actin promotor in clones

Wodarz stock collection

y1, w*, hsFlp; Sco/CyO

P{ry[+t7.2]=hsFLP}12, y[1] w[*]; sna[Sco]/CyO

Flipase under control of heat shock promotor; used for transdeletions

Bloomington #1929

P{XP}d00921 P{XP}CG43427d00921 P{XP} insertion at position 3R:485,651 [-]; used

for transdeletion of entire smash gene locus

Exelixis Collection,

Harvard

PBac{WH}f00542 PBac{WH}CG43427f00542 piggyBac insertion at position 3R:531,436 [+];

used for transdeletion of CG31534


Harvard

PBac{RB}e03181 PBac{RB}CG43437e03181 piggyBac insertion at position 3R:537,071 [-];

used for both transdeletions


Harvard

Df(3R)ED5066 w[1118]; Df(3R)ED5066,

P{w[+mW.Scer\FRT.hs3]=3'.RS5+3.3'}ED5066/TM6C, cu[1] Sb[1]

Deleted segment 82C5--82E4; deficient for smash gene locus

Bloomington #8092

smash4.1 smash[4.1] smash mutant allele; C-terminally truncated;

homozygous viable this work

smash35 smash[35] smash null allele; homozygous viable this work

Src64BKO Src64B[KO] Src64B nullallele O’Reilly et al.,

2006

Src42A26-1 Src42A[26-1] Src42A nullallele Takahashi et al.,

2005

Gla/CyO w[1118]; In(2LR)Gla, wg[Gla-1] Bc[1]/CyO Balancer for the 2nd chromosome Bloomington

#5439

If/CyO ; MKRS/TM6B

w[1118]; If/CyO; MKRS/TM6B[Tb1] Balancer for 2nd and 3rd chromosome Wodarz stock

collection

Br/CyO ; TM2/TM6B

w[*], Br/CyO ; TM2/TM6B[Tb1] Balancer for 2nd and 3rd chromosome Wodarz stock

collection

TM3/TM6B w[*]; TM3, Sb[1] Ser[1]/TM6B, Tb[1] Balancer line for the 3rd chromosome Bloomington

#2537

TM3[tw-GFP]/TM6B

w[*]; TM6B, Tb[1]/TM3, Sb[1], Ser[1], twist-GFP

Balancer line for the 3rd chromosome Wodarz stock

collection (B.Shilo)

UASt GFP-Smash-PI

UASt-GFP-Smash-PI GFP-Smash-PI under control of UASt promotor;

3rd chromosome Neugebauer,

2007

UASt GFP-Smash-PM

UASt GFP-Smash-PM[10] GFP-Smash-PMunder control of UASt promotor;

2nd chromosome this work

y1, w1118;; attp 22A3

y[1] w[1118]; PBac{y[+]-attP-3B}VK00037 attP docking site for phiC31 integrase-mediated

transformation; 22A3, 2L:1582820 Bloomington

#9752

UAS GFP-Smash-PI

UAS GFP-Smash-PI GFP-Smash-PI under control of UASp promotor;

2rd chromosome (22A3) this work

UAS GFP-Smash-PM

UAS GFP-Smash-PM GFP-Smash-PM under control of UASp

promotor; 2nd chromosome (22A3) this work

UAS Src42A-HA UAS Src42A-HA[11] Src42A-HA under control of UASp promotor, 2nd

chromosome this work


38

UAS Src64B-HA UAS Src64B-HA[6] Src64B-HA under control of UASp promotor, 2nd

chromosome this work

UAS Src42AYF-HA5

UAS Src42AYF-HA[5] Src42AYF-HA under control of UASp promotor,


UAS Src42AYF-HA6

UAS Src42AYF-HA[6] Src42AYF-HA under control of UASp promotor,


UAS mCD8::GFP w[*]; P{y[+t7.7] w[+mC]=10XUAS-

mCD8::GFP}attP2 mCD8-GFP under control of UAS promotor, 2nd

chromosome Bloomington

#32184

2.1.10 Fly breeding

Flies were raised on standard medium and kept at 25°C, unless stated otherwise. Embryo

collections were performed by keeping flies in cages on apple juice agar plates, which were

coated with yeast.

Standard medium consists of 356 g corn groats, 47.5 g soybean flour, 84 g dry yeast, 225 g malt

extract, 75 ml 10% nipagin, 22.5 ml propanoic acid, 28 g agar, 200 g sugar beet molasses and 4.9 l

H2O.

2.2 Genetic methods

2.2.1 Separation of DNA fragments via gel electrophoresis

DNA exhibits a negative charge associated with the phospho groups it contains. This allows DNA

to migrate in an electric field to the positively charged anode. 1% agarose gels containing

ethidiumbromide (EtBr), a chemical intercalating with DNA that can be visualized under UV light is

widely used to resolve DNA samples. The DNA is separated based upon its length and secondary

structure, which can be formed by circular DNA (e.g. plasmids). The size can be determined by

using a DNA standard, which is loaded in a separate lane on the gel.

1% agarose gel is made with 1 g agarose, which is solved in 100 ml of TAE buffer in a microwave. 1

µl of EtBr (1%) is added and the solution poured into a gel chamber. After hardening DNA can be

loaded and separation takes place at 110V for 20 – 25 min.


39

2.2.2 Polymerase Chain Reaction (PCR)

The polymerase chain reaction (PCR) is a method allowing the amplification of specific DNA

sequences, using primers (oligonucleotides) consisting of 18 – 30 nucleotides. The technique

became available when the thermophilic bacterium Thermophilus aquaticus was discovered. This

bacterium features a heat resistant DNA polymerase, afterwards called Taq polymerase. A

disadvantage of this enzyme is a lack of “proofreading” activity, leading to a significant error rate.

A second polymerase was isolated from the archae bacterium Pyrococcus furiosus, which features

a heat resistant DNA polymerase too. This enzyme exhibits a proofreading activity. This Pfu

polymerase is widely used for the cloning of DNA.

For the use of the PCR it is necessary to generate a pair of primers, which target the DNA

sequence of interest from 5’ to 3’ and 3’ to 5’, respectively. After denaturation at 95°C the DNA

will be available in a single stranded form. After decreasing the temperature, the DNA will

associate primarily with the added primers, because they are added in an excess. The

temperature for this annealing step is dependent on the sequence of the primer. Afterwards, the

DNA polymerase binds to the primer and a temperature shift to 72°C leads to the activation of the

enzyme and the subsequent synthesis of the complementary DNA strand by the use of dNTPs

(deoxynucleosidetriphosphates). Repeating these steps denaturation- annealing – elongation,

DNA can be easily amplified.

An example PCR can be set up as followed:

100 ng template DNA

4 µl oligonucleotides (10 pmol, forward and reverse)

2 µl dNTPs (10 mM)

10 µl polymerase buffer (5x)

1 µl polymerase

filled up to a volume of 50 µl with dH2O


40

Table 7 shows an example for a PCR program.

Table 7: Example for PCR program

step time temperature [°C]

I denaturation 5 min 95

II denaturation

30 sec 95

III annealing 30 sec dependent on sequence of the

respective primer

IV elongation

dependent on length of DNA sequence to be amplified

Taq polymerase 30 - 60 sec = 1 kb Pfu polymerase 90 sec = 1 kb

72

V repeat from step II for 34 times

VI final elongation

10 - 20 min 72

VII end of reaction

4

2.2.3 Mutagenesis PCR

Mutagenesis PCRs were adapted from the QuickChange II site directed mutagenesis kit

(Stratagene).

By the use of PCR it is possible to generate targeted mutations of DNA constructs. Point mutations

of single nucleotides can be achieved by designing primers, which are not fully complementary to

the template DNA. Thereby the coding nucleotide triplet is changed to generate the desired

mutation, meaning that during the annealing step these nucleotides will pair in a wrong way (e.g.

G with T or A). This triplet within the primer is flanked by 15 nucleotides on the 5’ and 3’ end. This

primer has a length of 33 nucleotides, which guarantees annealing of the oligonucleotide carrying

the desired base pair change.

Furthermore it is possible to delete whole domains, by “looping” them out. In this case the 5’

sequence of the primer is binding the template in front of the region, which is wished to be

deleted and the 3’ end binds after this region. In between this two 15 nucleotide long sequences

12 additional nucleotides can be added, which have a spacer function. The sequence can encode

for G – A – G – A. It is important that the deleted DNA construct stays in frame, thereby not

leading to a false translated protein. For this kind of PCR, the elongation time has to be adapted


41

to the length of the whole plasmid, because the polymerase has to amplify the full construct. The

PCR product needs to be digested by a restriction enzyme DpnI, which recognizes and digests only

methylated DNA (GAMTMC, which is modified in that way by E.coli). This step is necessary to get rid

of the non mutated template DNA. The mutated PCR product will be transformed into E.coli

DH5α.

2.2.4 Transformation of DNA into chemically competent E.coli

Chemically competent E.coli cells (e.g. XL 1 blue) are thawed on ice. 50 µl of bacteria are

transformed with DNA and incubation on ice takes place for 30 min. The cells will be heatshocked

for 45 sec at 42°C and incubated on ice again for 3 min. 250 µl of prewarmed SOC medium is given

to the cells which will be incubated at 37°C for 45 min while shaking. The mixture is plated on LB

agar plates containing an appropriate antibiotic for selection.

2.2.5 Isolation of DNA out of an agarose gel

After separating DNA by use of gel electrophoresis, bands can be cut out and the DNA isolated

and purified. For this purpose the NucleoSpin Extract II kit (Macherey Nagel) was used and

purification was according to the manufacturer protocol. DNA was eluted with 15 – 30 µl NE

Buffer.

2.2.6 Purification of DNA out of a PCR product

Purification was performed with the NucleoSpin Extract II kit (Macherey Nagel). DNA was eluted

with 15 – 25 µl of NE Buffer.


42

2.2.7 pENTRTM/D-TOPO® cloning

The principle of the pENTRTM/D-TOPO® cloning (Invitrogen) is to generate a pENTRTM/D-TOPO®

vector in the first step, which contains an insert of interest (e.g. coding sequence for CG43427-

RM). This vector can be used afterwards for Gateway® cloning to exchange the gateway cassette

from a destination vector with the insert of the pENTRTM/D-TOPO® vector. This reaction is

catalyzed by an enzyme called clonase, which recognizes attL sites upstream and downstream of

the pENTR insert/gateway cassette thereby mediating their recombination. Successful

recombination is selected by the change of antibiotic resistance. The pENTRTM/D-TOPO® vector

carries a resistance gene against kanamycin, and the destination vectors against ampicillin. Fig.10

shows the principle of the pENTRTM/D-TOPO® / Gateway® cloning system.

For the cloning as such it is necessary to design a forward primer with a CACC overlap at its 5’ end.

This signal will be recognized by the topoisomerase, which is included in the reaction mix with the

pENTRTM/D-TOPO® vector. For the reverse primer it has to be decided, whether one wants to

work with N-terminal or C-terminal tags. In the latter case the primer has to stop with the last

coding nucleotide triplet encoding for an amino acid. For N-terminal tags it is important to add a

nucleotide triplet encoding for a stop signal (e.g. TAA).

Fig.10: Principle of Gateway® cloning

An insert from the pENTRTM

/D-TOPO® vector is shuttled to a Gateway® destination vector, mediated by the

reversible reaction of the clonase. Selection is performed by the change of antibiotic resistence from

kanamycin to ampicillin (source: http://de-de.invitrogen.com/site/de/de/home/Products-and-

Services/Applications/Cloning/Gateway-Cloning/Clonase-Enzyme.html).


43

For the pENTRTM/D-TOPO® reaction, 2 µl of freshly purified PCR product were mixed with 0.5 µl of

salt solution and 0.5 µl of linearized pENTR vector. Incubation took place for 5 – 6 min at room

temperature. Transformation was done into the E.coli strain “top 10 one shot”. LB agar plates

containing kanamycin were used for selection.

The following clonase reaction was done with 50 - 150 ng of the pENTRTM/D-TOPO® vector, which

was mixed with 1.5 µl of TE buffer. 150 ng of the respective Gateway® destination vector was

added and reaction started with 1 µl of clonase. Incubation took place for 60 min at 25°C and

inactivation of the reaction was achieved by adding 0.5 µl of Proteinase K for 10 min at 37°C.

Transformation was done into E.coli DH5α or XL-1 blue. Constructs which were generated in this

work are listed in Table 8 below.

Table 8: Constructs generated in this work

name property

CG43427-PM pENTR

pENTRTM

/D-TOPO® vector containing the coding sequence of CG43427-RM

CG43427-PM pUASpGW

N-terminal tagged GFP-CG43427-RM under control of UASp promotor; contains attp site for site directed injection

CG43427-PM pTGW

N-terminal tagged GFP-CG43427-RM under control of UASt promotor

CG43427-PN pENTR

pENTRTM

/D-TOPO® vector containing the coding sequence of CG43427-RN

CG43427-PN pUASpGW

N-terminal tagged GFP-CG43427-RN under control of UASp promotor; contains attp site for site directed injection

CG43427-PE pENTR

pENTRTM

/D-TOPO® vector containing the coding sequence of CG43427-RE

CG43427-PE pUASpGW

N-terminal tagged GFP-CG43427-RE under control of UASp promotor; contains attp site for site directed injection

CG31534-PA Y64F pENTR

pENTRTM

/D-TOPO® vector containing the coding sequence of CG31534-RA carrying a pointmutation for Y64 to F

CG31534-PA Y64F pPGW

N-terminal tagged GFP-CG31534-RA with mutation for Y64 to F under control of UASp promotor


pENTRTM





pENTRTM





44


pENTRTM





pENTRTM





pENTRTM




CG31534-PA YmultiF pENTR

pENTRTM

/D-TOPO® vector containing the coding sequence of CG31534-RA carrying pointmutations for Y64, 152, 162, 244, 601 and 685 to F

CG31534-PA YmultiF pPGW

N-terminal tagged GFP-CG31534-RA with mutations for Y64, 152, 162, 244, 601 and 685 to F under control of UASp promotor

CG31534-PA Δprr1/2 pENTR

pENTRTM

/D-TOPO® vector containing the coding sequence of CG31534-RA were the two prolin rich regions were mutated

CG31534-PA Δprr1/2 pPGW

N-terminal tagged GFP-CG31534-RA with mutated proline rich regions 1 and 2 under control of UASp promotor

Src42A ΔSH3 pENTR

pENTRTM

/D-TOPO® vector containing the coding sequence Src42A where the SH3 domain was deleted

Src42A ΔSH3 pPWH

C-terminal tagged Src42A-HA where the SH3 domain was deleted, under control of UASp promotor

Src42A ΔSH2 pENTR

pENTRTM

/D-TOPO® vector containing the coding sequence Src42A where the SH2 domain was deleted

Src42A ΔSH2 pPWH

C-terminal tagged Src42A-HA where the SH2 domain was deleted, under control of UASp promotor

Src42A ΔKin pENTR

pENTRTM

/D-TOPO® vector containing the coding sequence Src42A where the tyrosine kinase domain was deleted

Src42A ΔKin pPWH

C-terminal tagged Src42A-HA where the tyrosine kinase domain was deleted, under control of UASp promotor

Src42A YF pENTR pENTR

TM/D-TOPO® vector containing the coding sequence Src42A where Y411 was

mutated to F

Src42A YF pPWH C-terminal tagged Src42A-HA where Y411 was mutated to F, under control of UASp

promotor


45

2.2.8 Isolation of DNA from bacteria

2 ml of LB medium containing the respective antibiotic were inoculated with a single E.coli colony

and grown over night at 37°C while shaking. 1.5 ml of the culture was transferred into a

microtube and bacteria were centrifuged for 1 min at 14.000 rpm. After discarding the

supernatant, the cell pellet was resuspended with 200 µl S1 resuspension buffer. 200 µl of S2 lysis

buffer were added and the mix vortexed. After this step cells were incubated for 5 min on ice.

Neutralization is achieved by adding 200 µl of S3 neutralization buffer. After this step the

microtube was inverted 5 – 6 times and subsequently centrifuged for 20 min at 14.000 rpm at 4°C.

The supernatant, containing the plasmid DNA, was transferred to a new tube. Precipitation of the

DNA was achieved by adding 400µl of isopropanol and centrifugation at 14.000 rpm at 4°C for 30

min. The supernatant was discarded and the DNA precipitate washed with 200 µl of ice cold 70%

ethanol and centrifuged for 5 min at 14.000 rpm at 4°C. Supernatant was discarded and the DNA

precipitate dried at room temperature. 25 µl of dH2O were given to the DNA for dissolving and

stored at -20°C.

2.2.9 Restriction digestion

By using restriction enzymes it is possible to cut DNA into fragments. These restriction

endonucleases can specifically cut DNA at the sequence motif they recognize, e.g. EcoRV is a type

II restriction endonuclease cutting GAT ↓ ATC into blunt ends (Pingoud and Jeltsch, 2001).

For a control digestion (e.g. after clonase reaction) 1-2 µg of DNA are digested in a volume of 10

µl, which contains 0.2 µl of the restriction enzyme and 1 µl of its respective 10x digestion buffer.

The restriction digestion takes place for 60 min at 37°C and DNA is afterwards separated by

agarose gel electrophoresis.

2.2.10 Determining DNA concentration

DNA concentration was determined with a photometer at OD260 (Eppendorf Biophotometer). The

dsDNA program was used with a dilution factor of 1:99. A small cuvette was filled with 99 µl dH2O


46

to define the blank value and then DNA concentration was obtained by adding 1 µl of the DNA

solution and a second measurement.

2.2.11 Sequencing of DNA

For the sequencing of plasmids 300 ng of DNA were used as template. For one reaction 1.5 µl of

SeqMix, 1.5 µl of SeqBuffer were set up with 8 pmol of a respective primer and the template.

Table 9 shows the PCR program, which was used for this reaction.

Table 9: Sequencing program

step time temperature [°C]

I denaturation 2 min 96

II denaturation 20 sec 96

III annealing 30 sec 55

IV elongation 4 min 60

V repeat from step II for 25 times

VI end of reaction 12

After the reaction the PCR product was transferred into a new microtube. Solution was mixed

with 1 µl 125 mM EDTA, 1 µl 3M sodium acetate (pH 5.2) as well as with 50 µl 100% Ethanol. After

incubation for 5 min at room temperature the microtube was centrifuged at 14.000 rpm for 15

min. The supernatant was discarded and the precipitate washed with 70 µl of 70% ethanol. After

centrifugation at 14.000 rpm for 5 min the supernatant was discarded, the precipitate air dried for

at least 2 min and resolved in 15 µl HiDi (formamide). Sequencing was performed by the

department of Prof. Dr. Pieler.

2.2.12 Isolation of genomic DNA from flies

Isolation of genomic DNA out of a single fly

By use of this method the isolation of genomic DNA out of a single fly is possible. For this purpose

a fly is put into a microtube and shock frozen in liquid nitrogen. Afterwards the fly was

homogenized with a biovortexer in 50 µl squishing buffer containing 0.5 µl Proteinase K.


47

Incubation for 30 min at 37°C was followed and inactivation of the Proteinase K was achieved by a

heat shock for 5 min at 94°C. Before PCRs have been set up, the lysate was centrifuged for 2 min

at 14.000 rpm to sediment the remaining cuticle of the fly. For a PCR of 25 µl volume 2.5 µl of the

supernatant were used as template.

Isolation of genomic DNA out of several flies (Quick fly genomic DNA prep)

30 flies were collected in a microtube and shock frozen in liquid nitrogen. Using a biovortexer flies

were homogenized in 200 µl buffer A, another 200 µl buffer A were added and further

homogenized till just cuticles were left over. The lysate was incubated for 30 min at 65°C and

subsequently mixed with 800 µl LiCl/KAc (575 µl 6 M LiCl + 230 µl 5 M KAc) solution. After

centrifuging at 13.000 rpm for 15 min the supernatant was mixed with 600 µl isopropanol in a

new miocrotube. Precipitation of the DNA was achieved after centrifuging for 15 min at 13.000

rpm. The supernatant was discarded and the DNA precipitate was washed with 200 µl of 70%

ethanol for 10 min at 13.000 rpm. DNA was resolved in 150 µl dH2O or TE buffer. This step was

done for 60 min at room temperature or at 4°C over night. The DNA was stored at -20°C.

2.2.13 UAS/Gal4 system

The UAS/Gal4 system is binary and was isolated from the yeast (Saccharomyces cerevisiae). The

Gal4 protein is the transcription factor and UAS (=upstream activating sequence) its binding site.

This system is used in Drosophila to mediate targeted expression of transgenes, which were

hooked up to an UAS element (Brand and Perrimon, 1993).

For this purpose a driver line, carrying the sequence for the Gal4 protein hooked up to a promoter

of a Drosophila gene (e.g. engrailed, en), is crossed against an UAS line. Hereby, expression is

controlled in temporal and spatial manner. en for example is a segment polarity gene, expressed

in stripes to the posterior compartment of segments. Therefore an en Gal4 driver line allows

expressing an UAS transgene in stripes, like the endogenous expression pattern of en. Fig.11

shows a schematic of the UAS/Gal4 system.


48

Fig.11: UAS/Gal4 system

The UAS/Gal4 system is used for the expression of transgenes in Drosophila. A driver line, expressing Gal4

under the control of any promoter, is crossed to a responder line, carrying a transgene fused to a the UAS

sequence. Adapted from Wimmer, 2003.

2.2.14 Generation of transgenic flies

By the use of P-elements and transposons it is possible to inject DNA into Drosophila embryos for

insertion into its genome (Bachmann and Knust, 2008).

The destination vectors of the Gateway® system are constructed with a P-element cassette and a

promoter (e.g. UASp or UASt) upstream of the coding sequence, as well as a mini white gene. It is

therefore suitable for the generation of transgenic fly lines. For the insertion it is necessary to co-

inject a helper plasmid (Δ2,3), encoding for a transposase, which recognizes the P-element and

mobilizes it and leads to its random insertion into the Drosophila genome.

For injection 20 µg of the respective DNA were mixed with 5 µg of Δ2,3 helper plasmid DNA. 5 µl

of 10x injection buffer (1 mM trisodium phosphate, 50 mM KCl, pH 6.8) were added and the

solution filled up with dH2O to a volume of 50 µl. Centrifugation for 10 min at 14.000 rpm at 4°C

was performed to get rid of particles, which could block the injection needle later on.

Injection was done into w1118 mutant embryos (these flies exhibit white eyes), which were laid for

20 min at 18°C. Embryos were transferred on a mesh with dH2O and dechorionized with sodium

hypochlorite for 1-2 min and subsequently aligned (to their anterior-posterior axis) on a piece of

apple juice agar. Afterwards embryos were transferred on a coverslip, which was coated with

embryo glue and dried at room temperature for 14 min (dependent on actual temperature and


49

humidity). During this step embryos lost their inner turgor, which prevents them from bursting

during the injection step. Right before injection embryos were covered with 10S oil and injected

into their posterior end where the pole cells will form later (the precursors of the germline). After

36-48 hours hatched larvae were collected and transferred into fly food. Emerging flies were

crossed against w1118; gla /CyO [ftz::lacZ] flies. In the second generation red eyed flies can appear,

which indicated the insertion of the transgene into their genome.

For the identification of the chromosome of insertion, a red eyed male fly either carrying the gla

marker or the CyO [ftz::lacZ] balancer chromosome has to be crossed against virgins of the w1118;

gla /CyO [ftz::lacZ] fly line. Table 10 shows the phenotypical markers for the identification of the

chromosome carrying the insertion (adapted and modified from Bachmann and Knust, 2008).

Table 10: Phenotypical markers for identification of the chromosome carrying the transgene

F2 w+ / gla F2 w

-; gla /CyO F2 w

+; gla /CyO F2w

-; CyO F2w

-; gla Chr.

male female male female male female male female male female

+ + + + + I

+ + + + II

+ + + + + + + + + + III

Another injection system used is based on the phi C31 integrase, allowing the insertion specifically

on landing positions (attp sites) in the genome (Groth et al., 2004). Here embryos were used for

injections which exhibit an attp site on position 22A on the second chromosome. Transgenes were

inserted by the phi C31 integrase, which is under control of the vasa promoter. An advantage of

this method is that established UAS lines display the same expression levels, thereby allowing

better comparison between the expression of different transgenes. The injection was performed

as described above, only the helper plasmid was not added to the injection mix. Furthermore,

emerging flies were crossed to w1118; gla /CyO [ftz::lacZ] and transformants were directly balanced

with CyO [ftz::lacZ].


50

2.3 Biochemical methods

2.3.1 SDS-PAGE

The SDS-PAGE (sodium dodecyl sulfate polyacrylamide gel electrophoresis) is a method which is

used to separate proteins according to their molecular weight. Hereby proteins are denaturated

and packed with SDS, giving them a total negative charge. In an electric field, these proteins will

migrate to the positively charged anode. The density of the gel can be controlled by the amount

of acrylamide which is used. For example, bigger proteins can be better separated by gels with

lower concentration of acrylamide, as in contrast smaller proteins by gels with higher

concentration. Table 11 lists the amount of contents for different gels.

Table 11: Examples of contents for different molecular SDS-PAGE gels

separation gel stacking gel

7.5% 10%

30% acrylamide 1.9 ml 2.5 ml 620 µl

Tris-HCl pH 8.8 2.8 ml 2.8 ml

Tris-HCl pH 6.8 470 µl

20% SDS 38 µl 38 µl 20 µl

dH2O 2.7 ml 2.1 ml 2.6 ml

10% APS 30 µl 30 µl 20 µl

TEMED 8 µl 8 µl 10 µl

SDS gels were made by using the BioRad system. Here two glass plates were fixed into a retaining

clip. The separation gel was poured in between the glass plates and covered with isopropanol.

After polymerization of the acrylamide the isopropanol layer was removed. Stacking gel was

poured on top of the separating gel and a comb, maintaining the space for the pockets, was

pushed between the two glass plates.

After polymerization of the stacking gel, the SDS gel was put into a BioRad electrophoresis

chamber (either two gels or one gel and one plastic dummy) and both chambers were filled with

1x SDS running buffer. Protein samples were loaded into the pockets with a syringe (Hamilton).

One pocket was filled with a protein mass ruler and protein separation took place for

approximately 60 min at 200V (depending on the concentration of polyacrylamide).


51

2.3.2 Western blot

For a Western blot, proteins are first separated via SDS-PAGE and afterwards transferred to a

nitrocellulose membrane. Detection of a specific protein occurs with an antibody and

concomitant signal development (e.g. chemiluminescence).

Horizontal transfer of the proteins from the SDS gel onto the nitrocellulose membrane occurs at

4°C for 60 min in transfer buffer using the BioRad system. Hereby a supplied blotting chamber is

assembled as follows: plastic mat – two whatman papers – SDS polyacrylamide gel – nitrocellulose

membrane – two whatman papers – plastic mat. After transfer the nitrocellulose membrane can

be stained with Ponceau red, which stains proteins and thereby indicates the quality of the

transfer. Subsequent incubation in blocking buffer is necessary to saturate unspecific binding

sites, which might cross react with antibodies. The nitrocellulose membrane had to be incubated

in blocking buffer containing a primary antibody in an appropriate dilution. This incubation was

done over night at 4°C.

After washing the nitrocellulose membrane 3 times for 5 min with TBST a secondary antibody

(conjugated with HRP) is added in blocking buffer at a dilution of 1:10.000. This incubation is done

for 2 hours at room temperature. Another 3 washes for 5 min followed with the subsequent

development of the chemiluminescence signal after incubating the nitrocellulose membrane for 1

min in POD solution (1ml solution A mixed with 10 µl solution B and 15 min incubation in the dark

before use, Roche). Detection of the signal was monitored using X-Ray developing films (Fuji).

2.3.3 Lysis of Drosophila embryos

For an embryonic protein extraction an overnight egg laying collection is usually used. Embryos

were washed with dH2O, dechorionized with sodium hypochlorite and transferred into a

microtube. Lysis was achieved by homogenizing the embryos in 200 µl lysis buffer (containing

proteinase inhibitors or additionally phosphatase inhibitors) with a biovortexer. The lysate was

filled up to a volume of 1000 µl and incubation on ice followed for 20 min. After a centrifugation

at 14.000 rpm at 4°C the supernatant was transferred into a new microtube and protein

concentration was determined. If protein lysates were not used directly, they were stored at -

20°C. Co-IPs were always continued without freezing to avoid disassembly of protein complexes.


52

2.3.4 Lysis of Drosophila S2 cells

S2 cells were harvested and centrifuged for 10 min at 1000 rpm. Culture medium was discarded

and cells were washed with 3 ml PBS. Another centrifugation at 1000 rpm for 5 min was done and

PBS removed. S2 cells were lysed by pipetting them up and down with 500 µl of lysis buffer

(containing proteinase inhibitors or additionally phosphatase inhibitors) and incubation on ice for

20 min. After centrifugation at 14.000 rpm at 4°C the supernatant was transferred into a new

microtube and protein concentration was determined. If protein lysates were not used directly,

they were stored at -20°C. Co-IPs were continued without freezing as mentioned above.

2.3.5 Determination of protein concentration

Protein concentration was determined with a photometer. 800 µl of dH2O were mixed with 200 µl

of Bradford reagent (Roth) and 2 µl of the used lysis buffer. This solution was incubated for 3 min

at room temperature. The blank value was set at OD600. Protein samples were also mixed in 800 µl

dH2O and 200 µl Bradford in the same way and the OD600 was measured. The value was multiplied

by 10 which then reflected the protein concentration in µg / µl.

2.3.6 Co-Immunoprecipitation (Co-IP)

An Immunoprecipitation (IP) is a method, where a specific protein is precipitated out of a protein

lysate. This is achieved by adding an antibody, which binds its antigen within this lysate. Under

physiological conditions (e.g. pH of lysis buffer) protein-protein interactions remain intact. It is

thereby possible to detect binding partners in Western blots after a Co-Immunoprecipitation (Co-

IP) (Wodarz, 2008).

For an IP it is important in which species the antibody was raised. The reason for this is that the

immunoglobulin domains of the antibodies show different affinities for the protein A/G

agarose/sepharose beads (see Table 12 for different affinities). Furthermore it is important that

the possible protein binding partner is bigger than 60 kDa in size or smaller than 50 kDa, if the

antibody against this protein is raised in the same animal species. The secondary antibody used


53

for Western blotting is targeted against the immunoglobulin domains of the animal species, which

does not allow to detect a protein of about 55 kDa, if the antibody used in the IP was from the

same animal species. In this case a possible interaction would not be detectable, because the

signal of the immunoglobulin domains would interfere.

Table 12: Affinities of protein A/G agarose for different immunoglobulins

species protein A protein B

rat + / - ++

mouse ++ ++

rabbit ++++ +++

guinea pig ++++ ++

For a Co-IP 1000 – 2000 µg of total protein were used in a volume of 500 – 1000 µl. An input

sample of 25 µg was taken for control. The lysate was preincubated with 20 µl of the respective

protein A or G agarose beads to get rid of unspecific binding proteins within the lysate. After

incubation at 4°C for 2 hours beads were centrifuged down for 1 min at 14.000 rpm at 4°C. The

supernatant was transferred into a new microtube and the respective antibody was added. As a

general rule 2 µl of serum antibodies were added or 1 µl of commercial ones (in the case of using

hybridoma supernatant antibodies, 15 µl of beads were preincubated in 300 µl of the respective

lysis buffer and 200 µl of the hybridoma antibody at 4°C for 2 hours; after centrifugation the

preincubated beads with the bound antibody were given to the lysate) and incubated over night

at 4°C.

15 µl of protein agarose beads were added and incubation at 4°C for 2 hours was carried out.

Beads were subsequently centrifuged down at 6.000 rpm for 30 sec at 4°C and washed three

times with the respective lysis buffer. The immunoprecipitated protein is coupled to the beads via

its bound antibody and interaction partners are also remaining bound to the precipitated protein,

if the physiological conditions were good. After denaturation of the protein agarose beads with 15

µl of 2x SDS loading buffer and concomitant separation via SDS-PAGE interaction partners can be

detected in Western Blot.


54

2.4 Cellculture

2.4.1 Transfection of S2 Schneider cells with FuGENE HD transfection reagent

Transfection for Co-IP experiments were done always with 2 wells for each (co-) transfection. Two

million cells were given into one well of a six well plate in 2 ml of S2 medium. Transfection mix for

one well was prepared as followed: 96 µl of dH2O containing 2 µg of the respective DNA

construct(s) were mixed with 4 µl of FuGENE HD transfection reagent. The mixture was vortexed

shortly and incubated for 15 min at room temperature. Transfection mix was given to the cells,

the six well plate carefully swung to achieve equal distribution of the solution within the well. S2

cells were incubated at 25°C for 48 hours and transferred to a cell culture flask, containing 5 ml of

fresh S2 medium. Cells were grown another 48 – 72 hours at 25°C before S2 cells were harvested.

2.5 Histology

2.5.1 Formaldehyde fixation of embryos

An overnight egg collection was transferred into a 15 mm Netwell (Corning Life Sciences, USA; 74

µm mesh size) and washed with dH2O. Netwell was placed into a supplied 12 well plate and

embryos were dechorionated in sodium hypochlorite for 1 – 2 min. After washing the embryos

with dH2O Netwell was held on top of a glass vial containing 4 ml heptane and was vigorously

inverted. After embryos were transferred into the glass vials, 3 ml of fixing solution (4%

formaldehyde in PBS) were added and fixation was performed for 18 min at room temperature

while rotating. The fixing solution (lower phase) was removed, 3 ml of methanol added and

embryos vortexed for 30 sec to remove the vitellin membrane. Embryos were transferred into a

microtube and washed with methanol twice. Storage could be done at -20°C or staining was

performed directly.


55

2.5.2 Formaldehyde fixation of larval tissue

L3 wandering stage larvae were selected and anaesthetized in PBS on ice. The posterior end was

cut off by using a micro scissor and the larva subsequently inverted on a needle. Larvae were

transferred into a microtube containing 885 µl PBS and 5 µl PBT. Fixation was started by adding

110 µl of 37% formaldehyde solution and was performed for 25 min on a rocking platform. Fixing

solution was removed and staining was continued (see 2.5.4).

2.5.3 Heat fixation of embryos

An overnight egg collection was washed with dH2O and dechorionized with 50% sodium

hypochlorite solution for 5 min. Embryos were washed on a mesh with dH2O and transferred with

a brush into a glass vial containing boiling 1x Triton salt. The vial was subsequently shaken for 5

times and put into ice and filled up with cold 1x Triton salt solution. Buffer was removed using a

plastic pipette and 3 ml of heptane and 3 ml of methanol were added. The glass vial was vortexed

for 30 sec to remove the vitelline membrane. Embryos were transferred into a microtube and

washed with methanol twice and were stored at -20°C for at least 2 hours before proceeding with

the staining.

2.5.4 Immunofluorescence staining

Fixed embryos were washed 3 times with PBT for 20 min at room temperature on a rocking

platform. PBT containing 5% NHS was given to the embryos for 30 min to block unspecific binding

sites. The primary antibodies were given to the embryos in PBT containing 5% NHS and incubation

took place at 4°C over night.

Embryos were washed 3 times with PBT for 20 min on a rocking platform. Secondary antibodies

coupled with fluorescence markers were given to the embryos in a concentration of 1:200 in PBT

containing 5% NHS. Incubation took place for 2 hours at room temperature. Afterwards embryos

were washed another 3 times for 20 min in PBT. During the first washing step DAPI was added in a


56

concentration of 1:1000 for marking the DNA. Embryos were mounted in a drop of Mowiol and

stored at 4°C.

Immunofluorescence staining of larval tissue was performed according to the protocol of embryo

staining. F-actin staining was achieved by use of fluorochrome conjugated phalloidin, which was

given to the secondary antibody solution in a dilution of 1:100. Phalloidin was placed into a fresh

microtube to evaporate the methanol in which it is dissolved. PBT containing 5% NHS and the

secondary antibodies were subsequently added.

2.5.5 Cuticle preparation

An overnight egg laying collection was taken and aged for another 24 hours to allow completion

of embryonic development. Hatched larvae were removed from the apple juice agar plate, as well

as embryos which did not possessed the respective genotype of interest (e.g. CyO [twist::GFP]/

CyO [twist::GFP]). Embryos were dechorionated in sodium hypochlorite, washed with dH2O on a

mesh and mounted with a drop of hoyers mountant. Incubation at 65°C was done overnight and

the coverslip was fixed afterwards with nail polish on the slide.

2.5.6 Wing preparation

Flies of the respective genotype were put into 100% isopropanol, wings dissected and laid on top

of a glass slide. After the isopropanol had evaporated, wings were mounted with a drop of Roti®-

Histokitt (Carl Roth) and a coverslip.

2.5.7 Eye preparation

Flies of the respective genotype were collected and placed into 100% ethanol in a glass dish.

Pictures of the eyes were taken from female flies with a Leica MZ 16 FA microscope and size


57

measurements were performed using the supplied scale bar from anterior to posterior as well as

from dorsal to ventral. Data were examined using Microsoft Excel.

RESULTS

58

3 Results

3.1 Baz binds to Smash in vivo

A yeast two-hybrid screen conducted by Ramrath, 2002 revealed that Baz might be a potential

interactor of the gene product of smash. To follow up on this finding, baits were used encoding

for either the region containing all three PDZ domains or for the regions containing PDZ 1 and 2

or PDZ 2 and 3 of Baz, respectively. The C-terminal part of Smash, exhibiting a LIM domain and a

class I PDZ binding motif (S/T X‡Φ§ -COOH) (Harris and Lim, 2001) was identified as prey (Fig.12).

Fig.12: Yeast two-hybrid screen with PDZ domains of Baz as bait

The gene product of smash was identified in a yeast two-hybrid screen as a potential interactor of Baz.

Bars and dashed bars indicate the region of the proteins, which were used in this screen. The PDZ

containing bait constructs were generated based upon the predicted PDZ domains at this time. In this

scheme, a current domain prediction by the smart server (http://smart.embl-heidelberg.de/) was used for

the PDZ domains. The oligomerization domain and the aPKC binding site are adapted from Benton and

Johnston, 2003. Furthermore, the short isoform Smash-PI as well as the larger isoform Smash-PM are

indicated.

http://smart.embl-heidelberg.de/

RESULTS

59

The subcellular localization shown in (Fig.13 A) shows colocalization of Smash and Baz in the

region of the AJs in embryonic ectodermal epithelium. A physical interaction was already shown

(Beati, 2009) but has been repeated in this work. Embryonic lysate was used for Co-IP

experiments, which expressed an N-terminally GFP tagged version of Smash-PI under control of

an UASt promotor with daughterless (da) Gal4. Baz was immunoprecipitated and GFP-Smash-PI

detected in Western blot (Fig.13 B). These results show that Baz and Smash are showing the

same subcellular localization in ectodermal epithelia and that they form a complex in vivo.

Fig.13: Smash colocalizes with Baz and binds to in vivo

(A) Staining for endogenous Smash protein (red) shows clear colocalization with Baz (green) in the region

of the AJs. The lateral membrane domain is marked with staining for Dlg (blue). DNA is stained with DAPI.

Scalebar = 10 μm. (B) da > GFP-Smash-PI. Co-IP experiments show that GFP-Smash-PI protein is detectable

by Western blot after IP of Baz from embryonic lysates. Co-IP experiment represents result of three

independent experiments.

3.2 Expression pattern of smash

To examine the expression of smash, two different antibodies were generated against

recombinant GST fusion proteins. One antibody was raised against an intramolecular region of

the short isoform Smash-PI (amino acids 328 – 634 of Smash-PI; 972 – 1278 of Smash-PM

respectively) in rabbit. This anti-Smash intra antibody showed apical staining of ectodermal

epithelia (Neugebauer, 2007), as well as expression in the trachea, where it strongly marks the

RESULTS

60

dorsal trunk and the dorsal branches (see Fig.14 B). However, this antibody does not work well

and harsh fixation methods are necessary (e.g. heat fixation).

While this work was in progress the gene annotation release 5.40 indicated that the neighboring

gene CG31531 located upstream of the CG31534 gene locus represents a single gene together

with CG31534, now annotated as CG43427 (see 1.5 and Fig.17 B). By PCR, using embryonic cDNA

as template, it was possible to amplify an overlapping fragment of both coding sequences,

confirming the reannotation of these two invidual genetic loci into a single molecular unit (data

not shown).

We therefore raised another antibody in guinea pigs against the first 300 amino acids of the very

N-terminus of the full length protein (see Fig.14 A).The anti-Smash N-term antibody works with

standard formaldehyde fixation and shows that Smash protein is detectable after cellularization

of the embryo present from stage 5 onwards throughout embryogenesis (see Fig.14 B). It is

localized cortically at the membrane in a honey comb pattern, similar to Baz. Cross section

images of the epithelium clearly show subcellular colocalization with Baz in the region were the

AJs are located (see Fig.15). The same subcellular localization was also observed when an N-

terminally HA tagged version of Smash-PI was expressed (Beati, 2009).

It was also apparent that Smash protein is strongly expressed in the amnioserosa in contrast to

Baz which is barely detectable there. Furthermore, Smash accumulated at the dorsal side of the

leading edge cells in a planar polarized fashion, where Baz is excluded from the cortex (see Fig.15

C and E and Laplante and Nilson, 2011). These results may indicate that membrane localization

of Smash is independent of Baz. Furthermore high levels of Smash were detected in the

developing hindgut, which was shown previously by RNA in situ hybridization (Ramrath, 2002).

RESULTS

61

RESULTS

62

Fig.14: Embryonic expression pattern of smash

(A) Scheme indicates regions of Smash which were used for antibody production. (B) Expression pattern

analysis using anti-Smash N-term antibody revealed that Smash is detectable throughout embryogenesis

after cellularization is completed. Note that from stage 14 onwards Smash protein accumulates in a stripe

like pattern, which reflects the structure of the denticle belts. The staining against Smash is specific, as

preimmune serum does not show signal (pre). Denticle belt staining is also observed with the anti-Smash

intra antibody. This antibody shows that Smash is present in the tracheal system, where it strongly marks

the dorsal trunk (dt) and the dorsal branches (db). Staining with this antibody is lost in the mutant

smash4.1

. Scalebar = 100 μm, embryos are positioned from anterior to posterior.

RESULTS

63

RESULTS

64

Fig.15: Subcellular localization of Smash

(A) Surface projection of embryonic epithelium showing membrane localization of Smash (red). (B) In a

cross section through the epithelium it is obvious that Smash is localized at the apical tip of the lateral

membrane, where the AJs are located and shows colocalization with Baz (green). Dlg was used as a marker

for the lateral membrane (blue). (C) High levels of Smash protein can be detected at the dorsal side of

leading edge cells (LE). Baz was barely detectable in the amnioserosa cells (as), whereas Smash shows

higher expression levels. (D) Smash localization at the AJs shown at a higher magnification. Smash is not

expressed in embryonic NBs (marked with asterisk). (E) Higher magnification of leading edge cells shows

that Baz is excluded from the dorsal membrane, whereas Smash accumulates dorsally in these cells.

Scalebars represent 20 µm in (A - C) and 10 µm in (D and E) respectively.

Using the UAS/Gal4 system, an N-terminal GFP tagged version of isoform Smash-PM, encoding

two additional coiled coil domains in its N-terminal part compared to the shorter isoform Smash-

PI (see 1.5 and Fig.12), could be shown to localize to the region of the AJs like the shorter

fragment. This shorter isoform also showed a comparatively stronger cytosolic localization

compared to the larger isoform (see Fig.16 A). Additionally, using an actin > CD2 > Gal4 flip out

line, GFP-Smash-PI was found in the nucleus of fat body cells, and the larger isoform GFP-Smash-

PM localized cortically as well as to a region adjacent to the nuclear envelope (see Fig.16 B).

However, the latter localization could not be shown with the antibodies raised against

endogenous Smash protein (data not shown).

RESULTS

65

RESULTS

66

Fig.16: Subcellular localization of Smash transgenes

Localization of N-terminally GFP tagged versions of Smash-PM and -PI. (A) Transgenes, carrying UASp

promoters inserted on the 2nd

Chromosome at position 22A, expressed with da Gal4. Both proteins had

been detected at the AJs, although GFP-Smash-PM showed much stronger apical localization compared to

the short isoform GFP-Smash-PI. Lower panels show higher magnification of subcellular localizations.

Scalebars represent 20 μm and 10 µm respectively. (B) Expression clones of N-terminally GFP tagged

versions of Smash-PM and -PI in fat body cells with an actin > CD2 > Gal4, UAS RFP flip out line. GFP-

Smash-PM was found at the cell cortex but showed localization to a region surrounding the nucleus as

well. However, the short isoform was strongly localized to the nucleus. Scalebar = 50 μm.

3.3 smallish knockout

In order to study the in vivo function of smash, a knockout allele was generated for the former

annotated gene CG31534 by transdeletion. Two FRT site containing piggyBac elements in trans

were recombined upon activation of a Flipase (Thibault et al., 2004). This led to deletion of the

genomic region present between both FRT sites. According to the reannotated genetic model,

this allele reflects a C-terminal truncation and therefore cannot be considered a classical null

allele, because it may still expresses the upstream region of the gene (see Fig.17 A). It therefore

could retain some function or furthermore cause dominant negative effects.

A full knockout allele was generated by a second transdeletion using two FRT-containing

piggyBac elements flanking the whole genomic region of smash (see Fig.17 B). Successful

deletions were confirmed by PCR (see Fig.18 B and C) and loss of antibody staining (see Fig.14

and Fig.20 A). Both mutations smash4.1 as well as the new allele smash35 are homozygous viable,

fertile and can be kept as homozygous stocks.

RESULTS

67

Fig.17: Transdeletion of the genomic locus of smallish

By using two different FRT containing piggyBac lines in trans it was possible to delete genomic DNA, which

was lying in between these FRT sites. (A) Genomic locus of the gene annotation releases 3.0 (2002) and 3.1

(2003). CG31531 is located upstream of the gene CG31534 (531,867 - 537,915). Position of piggyBac

elements is indicated (see Fig.18 for structure of piggyBac elements). Transdeletion led to a precise

knockout of CG31534. (B) Genomic locus according to the gene annotation release 5.40. smallish

(CG43427) reflects the genomic locus of the fusion between CG31531 and CG31534. Three new genes

were annotated, CR44156 and CR44157, as well as CG33927, which were also deleted by the new

knockout. P{XP} and piggyBac positions are indicated in the upper panel.

RESULTS

68

Fig.18: Verification of smash knockout

Mutant alleles for smash were verified by PCR on genomic DNA isolated from homozygous flies. (B)

smash4.1

and (C) smash35

. After transdeletion both alleles were screened for their ability to amplify DNA

corresponding to overlapping fragments from the piggyBac elements into the respective genomic regions.

(A) Scheme indicates overlapping PCR fragments with regard to the piggyBac elements (PCR fragments are

marked in red; size of piggyBac elements and PCR products does not reflect real length). Amplification of

both fragments corresponding to the overlapping fragments indicated successful transdeletion. This could

be shown for both mutant alleles. PCRs on genomic DNA of the deleted regions showed a loss of the

respective fragments (regions are not marked in the scheme). Structure of the piggyBac elements used is

shown in (A) and was taken from http://flypush.imgen.bcm.tmc.edu/pscreen/transposons.html. Direction

of insertion is indicated in the scheme with a [+] or [-] respectively.

http://flypush.imgen.bcm.tmc.edu/pscreen/transposons.html

RESULTS

69

Lethality was examined for F1 and F2 generations, as well as after shifting the temperature to

29°C from 25°C. One observation was that approximately 40% of mutant embryos die during

embryogenesis, either in the F1 or F2 generation. However, lethality after embryogenesis was

slightly increased in the F2 generation where about 25% of flies eclosed in contrast to

approximately 40% in F1. Shifting the temperature to 29°C increased lethality strongly. For the

individual lethality tests see Fig.19.

A

0

20

40

60

80

100

embryos (n)

hatched (n)

pupae (n) flies (n)

smash35 (F1)

0

20

40

60

80

100

embryos (n)

hatched (n)

pupae (n) flies (n)

smash35 (F2)

0

20

40

60

80

100

embryos (n)

hatched (n)

pupae (n) flies (n)

smash35 (F2 at 29°C)

RESULTS

70

Fig.19: Lethalities shown for smallish knockout

Lethality scores obtained for smash35

. (A) Lethality score for smash35

homozygous mutant embryos either

in the F1 or F2 generation and after temperature shift to 29°C. (B) Lethality score for the P-element lines

used for the transdeletion. (C) Complementation test of smash35

with deficiency line df(3R) ED5066

showing nearly the same lethality score as smash35

in (A). Data were obtained repeating the experiment

three times, error bars indicate standard error.

C

0

20

40

60

80

100

embryos (n)

hatched (n)

pupae (n) flies (n)

smash35 / df(3R) ED5066

B

0

20

40

60

80

100

embryos (n)

hatched (n)

pupae (n) flies (n)

P{XP}d00921 (F1)

0

20

40

60

80

100

embryos (n)

hatched (n)

pupae (n) flies (n)

PBac{RB}e03181 (F1)

RESULTS

71

Due to the fact that both mutant alleles generated for smash are homozygous viable,

mislocalization of its binding partner Baz was not excpected. Mutations in the baz gene reported

so far are lethal (see 1.1 and 1.5). Staining for Baz in embryos mutant for either smash35 or

smash4.1 showed that Baz is localized to the apical membrane. Furthermore, DE-Cad, an integral

component of the AJs, also showed normal localization (see Fig.20).

RESULTS

72

Fig.20: Epithelial integrity is not lost upon smash knockout

(A) Staining for Smash (anti-Smash N-term) is lost in the full knockout allele smash35

(shown in upper panel

in red), compare with Fig.13 A and Fig.15 B. Baz localization is not affected (green), lateral membrane is

marked by staining against Dlg (blue). DE-Cad protein localizes apically to the AJs (shown in lower panel in

red). (B) Staining for Baz (green) and DE-Cad (red) in the truncated knockout shows correct localization of

both proteins. Lateral membrane was marked with staining against Dlg (blue). (C) Higher magnification of

Baz and DE-Cad localization in smash35

mutant embryos. Scalebars represent 20 μm in (A and B) and 10 µm

in (C) respectively.

Examining protein levels of Baz, DE-Cad and Arm by Western Blot using embryonic lysates of

smash35 mutants could not show any change in their expression (see Fig.21). It was also tested

whether a change in protein levels could be observed when the N-terminal GFP tagged isoform

GFP-Smash-PM was overexpressed. No change in the expression levels of the respective protein

markers was detectable in Western blot.

RESULTS

73

RESULTS

74

Fig.21: Protein levels of AJ and polarity markers in smash mutants

To examine expression levels of AJ and polarity markers in smash mutants, protein lysates from smash35

homozygous mutant embryos were Western blotted and Baz, DE-Cad and Arm protein levels examined

using respective antibodies. (A) Embryo collection staged 2-4h AEL and (B) embryo collection staged 16-

18h AEL. As controls, piggyBac lines used for the transdeletion as well as white1118

embryos were taken for

comparison. Protein lysate from embryos overexpressing an N-terminal GFP tagged version of Smash-PM

using tub Gal4 was tested as well. No changes in protein levels were detectable. Anti-Smash N-term

antibody was not used, due to its background. This experiment was repeated three times, Actin is shown

as a loading control.

3.4 Overexpression of Smash

As shown in the previous chapter, smash gene function is not crucial for embryonic

development, nor for the survival of the adult fly (see Fig.19). In contrast to this finding,

overexpression using tub Gal4 (a strong driver line) and transgenic flies carrying transgenes for

N-terminal GFP tagged versions of Smash-PI or Smash-PM under control of UASt promotors

respectively, resulted in almost complete lethality. Overexpression of the short isoform GFP-

Smash-PI did not lead to embryonic lethality, but increased larval and pupal lethality (see Fig.22

A). Rare escaper flies that hatched were strongly reduced in size (see Fig.23 A and B), a result

also observed using da Gal4, but in a milder form (data not shown). In comparison,

overexpression of the larger isoform GFP-Smash-PM resulted in high embryonic lethality, where

almost 50% of embryos died before hatching (see Fig.22 B). Almost 25% of embryonic cuticles

displayed anterior holes, up to 5% dorsal holes and approximately 5% showed both anterior and

dorsal holes (see Fig.24). Hatched larvae died before pupation (see Fig.22 B). No adult flies

expressing GFP-Smash-PM under the control of tub Gal4 could be recovered. It was also tested

whether adult flies expressing GFP-Smash-PM under the control of tub Gal4 could be recovered

at 18°C, where the efficiency of the UAS/Gal4 system is reduced compared to 25°C, however no

escapers could be observed.

RESULTS

75

Fig.22: Lethality after overexpression of GFP-Smash epitopes

Lethality tests were conducted as mentioned before. tub Gal4 was used as a driver line and GFP positive

embryos were assayed. (A) Expression of an N-terminal GFP tagged short isoform Smash-PI leads to high

larval and pupal lethality. Rare escapers are observed that are reduced in size. (B) Expression of the

respective larger isoform Smash-PM, N-terminally tagged with GFP, leads to high levels of embryonic

lethality. Hatched larvae died before pupation. (C) CD8-GFP expression was used as control. All

experiments were repeated three times, error bars indicate the standard error.

A

0

20

40

60

80

100

embryos (n)

hatched (n)

pupae (n) flies (n)

tub > GFP-Smash-PI

B

0

20

40

60

80

100

embryos (n)

hatched (n)

pupae (n) flies (n)

tub > GFP-Smash-PM

C

0

20

40

60

80

100

embryos (n)

hatched (n)

pupae (n) flies (n)

tub > CD8-GFP (Bl32184)

RESULTS

76

Fig.23: Size decrease upon GFP-Smash-PI expression

(A) Rare eclosing escaper flies strongly expressing GFP-Smash-PI under the control of tub Gal4 show a

strong reduction in their size (upper female fly in comparison with a sibling carrying TM3 balancer

chromosome instead of tub Gal4). (B) Diagram showing the expression dependent size reduction of tub >

GFP-Smash-PI flies. The data represents measurements of flies from anterior to posterior which had been

raised at 18°C to obtain a sample size that was large enough to make a statistically relevant statement.

Too few escaper flies emerged at 25°C to make a statistically significant conclusion. Size was measured

with the supplied scale bar from Leica, error bars indicate standard error.

B

0

500

1000

1500

2000

2500

3000

males females

µm

TM3 / UASt GFP-smash-PI tub > GFP-Smash-PI

p < 0.01 ***

p < 0.01 ***

25°C 18°C

RESULTS

77

0

10

20

30

40

50

60

70

80

90

100

normal cuticle (n) anterior open (n) dorsal open (n) anterior + dorsal open (n)

morphology defects (n)

RESULTS

78

Fig.24: Cuticle phenotypes observed after overexpression of GFP-Smash-PM

(A) Embryos overexpressing GFP-Smash-PM under the control of tub Gal4 show variable cuticle

phenotypes. Approximately 50% of cuticles examined displayed no obvious defects. A subset of

approximately 25% showed an anterior hole phenotype, whereas 5% showed dorsal hole phenotypes.

Moreover, another 5% showed both anterior and dorsal hole phenotypes. In total, approximately 35% of

examined cultices showed combinations of these holes within the cuticle. Cuticles with other defects were

also observed which could not be classified and were summarized as morphology defects. (B) Diagram

showing statistical significance of phenotypes, which had been observed. Data were generated by

repeating the cuticle preparation three times, error bars indicate standard error.

Overexpression of the large isoform Smash-PM in a striped pattern using en Gal4 (see 2.2.13)

showed that these cells remained smaller in their size. The total length of AJs and the apical

surface area was significantly reduced (see Fig.26 B). Neighboring cells of GFP-Smash-PM

expression stripes were reduced in size compared to non-expressing cells of control embryos.

Smash might functions non cell autonomously and thereby shows an effect on neighboring cells

as well. Furthermore staining for DE-Cad showed that there is a slight accumulation of the

protein compared to non expressing cells (see Fig.25 A). This result indicates that Smash is

probably involved in pathways controlling apical constriction, which is a common feature of

epithelia undergoing morphogenesis (see 1.2). However, embryonic lysates from embryos

expressing GFP-Smash-PM under the control of tub Gal4 did not show any changes in total DE-

Cad levels (see Fig.21 B).

Expressing the same transgene in imaginal wing discs did not lead to malformed wings. Here, en

Gal4 was used to drive expression in the posterior half of the wing, or patched (ptc) Gal4, which

drives expression in an proximal/distal stripe (see Fig.27). dpp Gal4 was also tested, which

exhibits basically the same expression pattern as ptc Gal4 which also did not show any effect on

the wing shape (data not shown). Only wings from tub Gal4 driven overexpression showed an

overall size reduction which was expected given that the whole flies were reduced in size (see

Fig.23 A).

RESULTS

79

Fig.25: Overexpression of GFP-Smash-PM leads to cells smaller in size

Segmental expression of a UASt GFP-smash-PM transgene with en Gal4 leads to a decrease in cell size.

Expression of CD8-GFP in a striped pattern did not show any effect on cell size nor changes in protein

levels of DE-Cad or Baz in stage 11 embryos (B). However, expression of an N-terminally GFP tagged

version of Smash-PM caused those cells to remain smaller in size as compared to neighboring cells which

did not express the transgene. Interestingly DE-Cad levels at the membrane appeared to be slightly

increased (A). Scalebars =100 μm in the overviews and 10 µm in the respective magnification.

RESULTS

80

Fig.26: AJ length and apical surface area is reduced upon GFP-Smash-PM expression

Total length of AJs and apical membrane area was analyzed in en Gal4 expression stripes. Area close to the

end of the elongated germband of stage 11 embryos was chosen for analysis. (A) Example of how the

length of AJs and the apical membrane area was determined. DE-Cad staining (blue and white in merge)

was used to mark the AJs (red marked cell). LSM software provided respective AJs length in µm and apical

surface area in µm2. Scalebar = 10 µm. (B) AJs length and apical surface area is comparable in non

expressing cells and en Gal4 expressing CD8-GFP stripes. A significant reduction in AJs length was observed

upon expression of GFP-Smash-PM. Apical surface area was strongly reduced compared to non expressing

neighboring cells. However, non expressing cells of control embryos showed that AJs were longer as

compared to non expressing cells of GFP-Smash-PM expressing embryos. Comparable observation was

also made with regards to the apical surface area. This might indicate that Smash is functioning non cell

autonomously. Error bars indicate standard error.

0,00

5,00

10,00

15,00

20,00

25,00

[µm

]

AJ length

0,00

5,00

10,00

15,00

20,00

25,00

[µm

2 ]

apical surface area

p < 0.01 ***

p < 0.01 ***

p < 0.01 ***

p < 0.01 ***

RESULTS

81

Fig.27: Expression of GFP-Smash-PM does not show an effect on wing shape

Using en Gal4 as a driver allows specific expression of transgenes in the posterior compartment of the

developing wing. (A) Expression of the N-terminal GFP tagged short isoform Smash-PI or the larger protein

Smash-PM do not show effects on the wing shape. As control UAS CD8-GFP (Bl 32184) was used. (B) ptc

Gal4 used as a driver line, which expresses in a stripe from proximal to distal in the developing wing. No

wing malformation had been observed. Scalebars = 500 μm.

Overexpression of these transgenes in the eye using pGMR Gal4 as a driver resulted in a rough

eye phenotype (see Fig.28 A). Eyes were furthermore reduced in their size in comparison to the

transgenic lines without expression (see Fig.28 B). The rough eye phenotype was slightly

enhanced by expressing two copies of the large isoform GFP-Smash-PM. The rough eye

phenotype observed for Smash overexpression was also slightly apparent in the rare escapers

using tub Gal4 (data not shown).

RESULTS

82

Fig.28: Eye restricted expression of Smash leads to rough eyes and size reduction

pGMR Gal4 driven expression of N-terminal GFP tagged versions of Smash. (A) Figures show results of

expression of GFP-Smash-PI as well as GFP-Smash-PM in the eye. Upper panel shows respective transgenic

lines and pGMR Gal4. UASt GFP-smash-PM line was recombined with pGMR Gal4 and crossed against the

UASt GFP-smash-PM line for expression of two copies of the transgene, which led to a slightly enhanced

phenotype. (B) Diagram summarizes results of eye sizes. Measurements were performed for the

anterior/posterior axis (a/p) as well as for the dorso/ventral axis (d/v) by using the supplied scalebar from

Leica. Expression of either GFP-Smash-PI or GFP-Smash-PM led to a size reduction in the a/p axis,

compared to pGMR Gal4 and the transgenic lines. Error bars indicate standard error.

0

100

200

300

400

500

600

700

[µm

]

a/p d/v

pGMR Gal4 / CyO (n=10)

UASt GFP-Smash-PI (n=8)

UASt GFP-Smash-PM (n=10)

pGMR > GFP-Smash-PI (n=8)

pGMR > GFP-Smash-PM (n=4)

p < 0.02 **

p < 0.01 ***

p < 0.01 ***

RESULTS

83

3.5 Smash binds Src42A and Src64B in vivo

In order to get more insight into the gene function of smash, other potential binding partners of

Smash were examined. A yeast two-hybrid screen which was designed for an interaction map of

Drosophila proteins (Giot et al., 2003), indicated that the non-receptor tyrosine kinase Src42A is

a potential binding partner of Smash-PI and Smash-PJ, the latter representing a slightly shorter

isoform lacking the LIM domain by an alternatively spliced exon containing a stop signal (see 1.5

and Fig.32, prior annotation CG31534-PB). It was previously shown that Src42A and Src64B have

the abilities to bind Smash-PI (Beati, 2009), but additional effort was investigated in this work.

Antibody staining against endogenous Src42A protein showed that it is localizing along the

basolateral membrane of epithelial cells but showing slightly higher accumulation in the region

of the AJs (see Fig.29 A and 1.3). Furthermore a phosphospecific antibody directed against

tyrosine phosphorylated Src42A, which represents an activated form, is localizing specifically at

the AJs (see Fig.29 B and 1.3) and at places where morphogenetic processes like the invagination

of the cephalic furrow happen (Shindo et al., 2008). As it has been proposed and supported by

antibody stainings that the region of AJs are rich in phosphotyrosines (see Fig.29 C and 1.3)

Src42A was a good candidate for further studies.

RESULTS

84

RESULTS

85

Fig.29: Subcellular localization of Src42A and the activated form pSrc

Stainings show subcellular localization of Src42A and pSrc in white1118

embryos, as well as in mutants for

smash4.1

. (A) Src42A was detected along the basolateral membrane with slightly higher accumulation at

the apical membrane. Arm is shown as an AJs marker. Merged image shows Src42A in red and Arm in blue.

Lower panel shows subcellular localization of Src42A and Arm in smash4.1

mutant embryos. (B) Staining for

the activated form of Src42A, labeled as pSrc (Shindo et al., 2008), showed strong localization in the region

of the AJs. Lower panel shows no apparent mislocalization of pSrc in smash4.1

mutants. (C) Staining for

phosphotyrosine (YP) shows that AJs are rich in proteins with phosphorylated tyrosines. Baz was stained as

an AJs marker. Phosphotyrosine levels were not affected in smash4.1

mutant embryos. Scalebars = 10 μm.

To test whether both proteins interact physically, N-terminally GFP tagged Smash-PI was co-

expressed in S2 cells either with C-terminally HA tagged Src42A or Src64B, respectively. After

immunoprecipitation of GFP, HA was detected by Western blot, indicating that GFP-Smash-PI

binds to both Src kinases in vitro in this cell culture system. Furthermore, using an antibody

against phosphotyrosine, a strong phosphorylation signal was detected at the respective

molecular weight of GFP-Smash-PI, which was not observed after expression of GFP-Smash-PI

without Src42A (see Fig.30 A). This finding strongly indicates that Smash-PI is phosphorylated by

Src42A in vitro.

Smash was also shown to be tyrosine phosphorylated in vivo. Embryonic lysates were prepared

from white1118 and smash4.1 mutants using phosphatase inhibitors in the lysis buffers. Smash was

immunoprecipitated using the anti-Smash intra antibody and phosphotyrosine antibody was

used for Western blot. A phosphorylation signal could be observed for white1118 but not for

smash4.1 mutants (see Fig.30 B). This gives additional evidence to the identified interaction.

RESULTS

86

RESULTS

87

Fig.30: GFP-Smash binds to Srcs and is tyrosine phosphorylated in vivo

(A) Expressing an N-terminal GFP tagged version of Smash-PI together with C-terminal HA tagged versions

of Src42A or Src64B respectively, showed that both non-receptor tyrosine kinases were detectable in

Western Blot after immunoprecipitation of GFP-Smash-PI. Furthermore, using an antibody directed against

phosphotyrosine, a corresponding signal was observed at a molecular size of GFP-Smash-PI upon co-

expression with Src42A-HA. This phosphorylation event appeared to specifically require Src42A-HA as co-

expression with Src64B-HA resulted in very little phosphorylation. (B) The tyrosine phosphorylation of

Smash was also shown in vivo. white1118

embryonic protein lysates were compared with lysates from

homozygous smash4.1

mutants for phosphotyrosine levels. Smash was immunoprecipitated with anti-

Smash intra antibody and concomitant Western blotting for phosphotyrosine showed a phosphorylation

signal, which was lost in the mutant. (C) Phosphotyrosine levels of Arm were unaffected in mutants for

smash4.1

in comparison to white1118

embryos.

It has been suggested that LIM domain containing scaffolding proteins could act as adapters

between kinases and their respective targets (Khurana et al., 2002). Due to the fact that Arm is a

phosphorylation target of Src42A and Src64B (Takahashi et al., 2005), its tyrosine

phosphorylation state was analyzed in mutants for smash4.1. Embryonic lysates of the C-terminal

truncated allele smash4.1 were used, Arm was immunoprecipitated and phosphotyrosine signal

was analyzed in Western blot. A change in phosphorylation was not observed in mutants for

smash4.1 (see Fig.30 C).

In an attempt to uncover protein domains required for the interaction between Smash-PI and

Src42A a set of domain deletions were generated. Either the N-terminal SH3 domain, the SH2

domain or the C-terminal tyrosine kinase domain were deleted (see Fig.31 A). However, none of

these deletions showed complete abolishment of the interaction with GFP-Smash-PI (see Fig.31

B). Furthermore, deletion mutants lacking the SH3 or SH2 domain were still able to

phosphorylate GFP-Smash-PI. Only deletion of the tyrosine kinase domain showed loss of

phosphorylation, as expected. However, Src42A ΔSH2-HA showed a reduced ability to

immunoprecipiate with GFP-Smash-PI, possibly indicating a special role for this domain in the

context of this interaction.

RESULTS

88

Fig.31: Src deletion Co-IPs

(A) Different Src42A domain deletions generated with the C-terminal HA tag shown in turquoise. Upper

panel shows the N-terminal GFP tagged version of the short isoform Smash-PI. Both proline rich regions

(prr) are indicated. (B) Co-IP experiments showed that all Src42A deletions generated can still bind to GFP-

Smash-PI. Phosphorylation of GFP-Smash-PI was still detected after SH3 or SH2 deletion. A slight decrease

in binding was observed upon SH2 deletion. Vice versa mutations in both proline rich regions of Smash-PI

did not show any decrease in the binding ability between both proteins. Mutations in the proline rich

regions of Smash-PI showed no effect on Src42A binding.

RESULTS

89

The tyrosine phosphorylation shown in cell culture experiments (see Fig.30 A and Fig.31 B) as

well as in vivo in embryos (see Fig.30 B) raised interest in the phosphorylation target sites.

Phosphorylation of the protein by Src42A may have functions such as activation or inactivation

and overexpression of respective point mutated forms of the protein could have shed light on its

biological function.

The NetphosK software from the technical university of Denmark

(http://www.cbs.dtu.dk/services/NetPhosK/) was used to identify Src specific phosphorylation

motifs within the amino acid sequence of Smash-PI. Five tyrosine residues were detected as Src

phosphorylation target sites (Y64/708, Y152/796, Y162/806, Y244/888 and Y601/1245, red

numbers indicate respective sites in Smash-PM). An additional residue (Y685/1329) was

identified by PhosphoPep, a phosphoproteome resource of Drosophila proteins based on mass

spectrometry (Bodenmiller et al., 2007). Due to the change in the gene annotation release a

further tyrosine residue was detected in the N-terminal region of Smash-PM by the NetphosK

software as a Src motif, which is Y295. This residue has not been investigated yet.

The identified tyrosine residues were mutated to phenylalanine and co-expressed with Src42A-

HA in S2 cells to monitor the phosphorylation state of each respective point mutated protein.

Y64/708 as well as Y162/806 showed decreased phoyphorylation signal, which was supported by

loss of one band (a double band was usually detectable after separation with lower percentage

SDS-gels) detected by Western blotting and probing against phosphotyrosine (see Fig.32).

Furthermore a mutant form carrying mutations in all of the identified tyrosine residues showed

strongly reduced phosphorylation but not a complete abolishment. These results indicate that

the tyrosine residues, Y64/708 and Y162/806, most likely represent two phosphorylation sites

for Src42A. However, there must be at least one more target site, which was not detected by the

NetphosK prediction software.

http://www.cbs.dtu.dk/services/NetPhosK/

RESULTS

90

Fig.32: Analysis of Smash-PI phosphomutants

In order to unravel the Src phosphorylation target sites, a set of point mutations were generated for

Smash-PI. (A) Scheme indicates the Src phosphorylation motifs by the NetphosK prediction server of the

Technical University of Denmark and the single tyrosine residue 685 predicted by the PhosphoPep analysis

(Bodenmiller et al., 2007). Upper panel shows the respective sites in the short isoform Smash-PI and

Smash-PJ, which lacks the LIM domain due to alternative splicing. The corresponding sites are indicated for

the larger isoform Smash-PM. An additional tyrosine was predicted by NetphosK in the N-terminal region.

(B) Co-IP experiments between Src42A-HA and different point mutated forms of N-terminal GFP tagged

Smash-PI. Predicted tyrosines were exchanged to phenylalanine. Western blot and probing for

RESULTS

91

phosphotyrosine implicated Y64 and Y162 as phosphorylation targets because the phosphorylation signal

is strongly decreased. Moreover a respective mutant form, carrying mutations to phenylalanine in all

predicted sites, showed almost complete abolition of the phosphorylation signal.

3.6 Overexpression of Src42A in the eye

It was previously reported that overexpression of SFK transgenes in the Drosophila eye, as well

as their constitutively active forms, causes interruption of normal development (Pedraza et al.,

2004). Based on this assay different transgenic fly lines were generated with constructs encoding

C-terminally HA tagged forms of Src42A, Src64B and two different alleles for constitutively active

forms of Src42A (Y511 to F mutation). These cannot be phosphorylated by Csk anymore (see

1.3). Expression of these transgenes with pGMR Gal4 reproduced the reported phenotypes,

although they were milder here (see Fig.33 A). Since SFKs are involved in cell proliferation

control another observation was a change in eye size. As expected, the expression of the

constitutively active forms showed slightly stronger phenotypes.

It has been published that the vertebrate homolog of smash, LMO7 (see 1.4) has tumor

suppressor functions in mice (Tanaka-Okamoto et al., 2009). To check whether smash could also

have tumor suppressor functions especially with regard to its interaction with Src42A, expression

of the above mentioned Src transgenes were performed together with GFP-Smash-PM. As

already shown in Fig.28 A, expression of an UASt GFP-smash-PM transgene causes a rough eye

phenotype as well as size reduction along the a/p axis (see Fig.28 B). However, co-expression

with Src42A YF-HA led to an enhancement of the rough eye phenotype and a reduction in size.

Vice versa, expressing these Src transgenes in eyes mutant for smash would be of great interest.

RESULTS

92

C

0

100

200

300

400

500

600

700

a/p d/v

[µm

]


UAS Src42A-HA (n=6)

pGMR > Src42A-HA (n=4)

UAS Src64b-HA (n=5)

pGMR > Src64B-HA (n=6)

D

0

100

200

300

400

500

600

700

a/p d/v

[µm

]


UAS Src42A YF5-HA (n=5)

pGMR > Src42A YF5-HA (n=6)

UAS Src42A YF6-HA (n=9)


p < 0.01 ***

p < 0.01 ***

p < 0.05 *

p < 0.01 ***

p < 0.01 ***

p < 0.05 *

RESULTS

93

Fig.33: Expression of Src in the eye

pGMR Gal4 driven expression of different Src transgenes. (A) Rough eye phenotypes caused by expression

of Srcs. Slight rough eye phenotypes have been observed upon expression of either Src42A-HA or Src64B-

HA. Size measurements revealed a slight increase along the a/p as well as the d/v axis (C). Phenotypes

were enhanced by expressing constitutively active forms of Src42A (carrying Y511F mutation). The eye size

was increased as well compared to the non mutated form of Src42A (D). (B) The rough eye phenotype was

slightly enhanced by co-expressing the constitutively active forms of Src42A together with GFP-Smash-PM.

Furthermore, the enlarged eye phenotype was suppressed slightly in the a/p axis but not in the d/v axis

(E). Error bars indicate standard error.

3.7 smallish genetically interacts with Src64B

Single mutants for Src42A as well as Src64B do not show strong defects with regards to

morphogenetic processes like dorsal closure. However double mutants for both kinases exhibit

severe defects in dorsal closure and head involution (see 1.3). To examine whether smash has

redundant functionality in these pathways, double mutants for both kinases were generated.

Src42A single mutants are homozygous lethal but Src64B mutants are homozygous viable. The

Src42A26-1 mutation, the strongest reported mutant allele for Src42A (Takahashi et al., 2005), can

still be kept in the homozygous mutant background of smash4.1. This observation indicates that

both proteins act in the same pathway, as one copy of Src42A is still enough for survival in the

smash mutant background.

E

0

100

200

300

400

500

600

700 [µ

m]

a/p d/v


UASt GFP-smash-PM (n=10)

pGMR > GFP-Smash-PM (n=4)


pGMR, Src42A YF5-HA / GFP-Smash-PM (n=5)


pGMR, Src42A YF6-HA / GFP-Smash-PM (n=6)

p < 0.01 ***

p < 0.01 ***

RESULTS

94

Interestingly a double mutant for smash4.1 and Src64BKO shows high embryonic lethality (see

Fig.34 A and B and Fig.19 A for smash35 lethality scores). Approximately 70% die during

embryogenesis and larvae furthermore die immediately after hatching. However, some eclosing

escapers were observed. Embryonic cuticles do not exhibit dorsal closure defects (see Fig.35 A),

which were reported for Src42A and Src64B double mutants (see 1.3). Stainings for Baz and DE-

Cad did not show mislocalization, indicating that cell polarity and the formation of AJs remains

intact, which supports the finding that some escapers were observed. However, a da Gal4 driven

transgene of GFP-Smash-PM could not significantly rescue the lethality of this double mutation,

although a few homozygous flies were recovered. smash may exhibit more isoform specific

functions (10 isoforms are annotated on flybase), which may explain this finding. Unfortunately a

resuce using transgenic flies carrying the genomic locus of smash could not be performed

because the injected genomic clone did not give rise to any transformant flies. A rescue

experiment using a transgene for Src64B has not been tested so far. These results may indicate a

new redundant pathway for smash in regard to Src64B which is not important for dorsal closure.

Fig.34: Lethality of smash and Src64B double knockout

Double-knockout of Src64B and smash showed significantly increased lethality. (A) The null allele Src64BKO

is homozygous viable and can be kept as a homozygous stock. The same observation was made for both

smash alleles (see Fig.19 A for lethality scores of smash35

). (B) A double-knockout of smash4.1

and Src64BKO

showed high embryonic lethality. Almost 70% died during embryogenesis, hatched larvae usually died

immediately. Very few eclosing escapers were observed. Lethality scores represent data from three

experiments, error bars indicate standard error.

A

0

20

40

60

80

100

embryos (n)

hatched (n)

pupae (n) flies (n)

Src64BKO

B

0

20

40

60

80

100

embryos (n)

hatched (n)

pupae (n) flies (n)

smash4.1 , Src64BKO

RESULTS

95

RESULTS

96

Fig.35: Double-knockout of smash and Src64B shows no dorsal closure defects and normal epithelial

integrity

(A) A double-knockout for smash and Src64B does not show cuticular defects. (B) Heterozygous embryo

compared to a homozygous double mutant for smash and Src64B, marked by GFP expression of the TM3

balancer chromosome. Difference in Baz and DE-Cad levels were not detectable. Dorsal closure was

completed in the homozygous mutant. Scalebar = 100 µm. (C) Epithelia showed normal organization.

Staining for Baz showed correct apical localization, DE-Cad was found to be localized at the AJs. These

findings indicate that lethality of the double mutant is probably not caused by defects in epithelial cell

polarity. Scalebars represent 20 μm in the surface projections and 10 µm in the cross sections respectively.

DISCUSSION

97

4 Discussion

In this study we investigated the function of the so far undescribed gene CG43427, which we

named smallish (smash) due to its reduced size phenotype caused by overexpression. The gene

product of smash was identified in a yeast two-hybrid screen as a potential interactor of Baz

(Ramrath, 2002), a keyplayer in regard to cell polarity and AJs formation (Johnson and Wodarz,

2003; McGill et al., 2009). In this screen the three PDZ domains of Baz were selected as bait.

These domains are known to be protein-protein-interacting modules (Sheng and Sala, 2001; Te

Velthuis and Bagowski, 2007). The C-terminus of Smash possesses a PDZ binding motif of class I

(S/T X‡Φ§ -COOH) (Harris and Lim, 2001), a motif binding to PDZ domains. Furthermore, a

vertebrate homolog of smash, LMO7, had already been shown to function at AJs (Ooshio et al.,

2004). We confirmed the in vivo binding of Baz and Smash and continued studying the

developmental relevance of this interaction. Furthermore we showed that the non-receptor

tyrosine kinase Src42A is a binding partner of Smash in vitro (Beati, 2009). We further investigated

this interaction and found that Src42A phosphorylates Smash and that smash genetically interacts

with the second Src kinase encoded by the Drosophila genome, Src64B.

4.1 Baz binds to Smash in vivo

As we have shown previously (Beati, 2009), Baz forms a complex with Smash in vivo in embryos.

At this time we did not have access to antibodies against Smash that worked well in Western

blots. The anti-Smash intra antibody showed many unspecific bands. In this work we used

embryonic lysates expressing an N-terminally GFP tagged version of Smash-PI, which showed an

ability to co-immunoprecipitate with Baz. However, it would be interesting to test whether

endogenous Smash can be detected as well, when Baz is precipitated from embryonic lysates, as

we do not know anything about the stoichiometry of this binding. Furthermore it was only tested

whether Baz forms a complex with Smash during embryogenesis. It would be interesting to test

whether Baz and Smash can be still detected in a complex in vivo in larval as well as in adult

tissues. Since Smash was identified as a potential binding partner of Baz by a yeast two-hybrid

screen, the likelihood is high that both proteins can bind directly to each other, which we have

not tested so far. Further evidence for direct binding of Baz to Smash is given by additional yeast

DISCUSSION

98

two-hybrid analyses, where only PDZ 1 and PDZ 2, or PDZ 2 and PDZ 3 were used as baits

respectively (Ramrath, 2002; see Fig.12). In both cases, the C-terminus of Smash still bound to the

bait proteins. GST pulldown assays could show whether both proteins can interact directly, and if

so, which of those three PDZ domains is of importance by using PDZ deletion versions of Baz.

As we found that Baz is colocalizing with Smash in the region of AJs, it would be important to

analyze whether Smash mislocalizes in mutants for baz. So far we have not investigated whether

Smash is expressed in the follicular epithelium, which is derived from the mesoderm. An easy test

would be to induce baz mutant clones in this tissue and co-stain for Smash. However, a baz

deletion allele is not available and respective mutants may still express defective fragments of Baz

(Shahab, unpublished data). However, some alleles are supposed to lack the region carrying the

PDZ domains and thus could be used for the respective experiment (Krahn et al., 2010b). It was

reported that Baz is excluded dorsally in leading edge cells in a planar polarized fashion (Laplante

and Nilson, 2011). As we detected that Smash shows strong accumulation dorsally in leading edge

cells (see Fig.15) it may indicate that Smash can localize to the plasma membrane in a Baz

independent way. Furthermore staining for Baz showed low protein levels in the amnioserosa,

whereas Smash was clearly detectable in this tissue. Furthermore, Smash transgenes lacking C-

terminal parts still localized to the membrane (Beati, 2009). Currently, all generated transgenic fly

lines were created with the old injection system. It would be important to test the localization of

respective deletion mutants with site directed insertion of the respective transgenes. This would

guarantee comparable expression levels via the UAS/Gal4 system. Another mechanism for Smash

localization could be via Src42A. However, Smash localizes to regions containing AJs, whereas

Src42A localizes along the whole plasma membrane (see Fig.29). It would therefore be interesting

to analyze this interaction with regards to pSrc, which was shown to exclusively localize at AJs

(Shindo et al., 2008). It is possible that Smash may only bind to pSrc, however, we could not

discuss this specific interaction using our experimental setup.

DISCUSSION

99

4.2 Expression of smallish

As the anti-Smash intra antibody did not work well with standard fixation methods (e.g.

formaldehyde fixation) and staining required harsh fixation methods (e.g. heat fixation), co-

stainings with other proteins were not easy. Due to the change in the current gene annotation

release 5.40 and the fact that smash consists of both transcription units of CG31534 and CG31531

we decided to generate a second antibody. As shown in Fig.14 we could show ubiquitous

expression of Smash in ectodermal epithelia throughout embryogenesis. Staining is apparent after

cellularization of the embryo, when the first epithelium is established (Knust and Bossinger,

2002). Currently it is not clear if Smash is maternally deposited into the egg in the form of

proteins or mRNAs. To determine when zygotic expression begins, female mutants for smash

could be crossed to wildtype males. Staining for Smash would indicate the start of zygotic

expression, since any maternal supply would be depleted in those eggs.

We found that Smash protein shows similar subcellular localization as Baz. It localizes at the

membrane in a honey comb fashion and was found to localize at AJs (see Fig.15). Immuno

electron microscopy could reveal the subcellular localization in greater detail. The vertebrate

homolog LMO7 was shown to localize specifically at AJs (see 1.4) and at the free apical membrane

(Ooshio et al., 2004).

We further detected Smash in the tracheal tubes using the anti-Smash intra antibody. This

staining was shown after heat fixation and was lost in the smash4.1 mutant allele, supporting the

specifity of the staining. However, the anti-Smash N-term antibody only showed non specific

tracheal staining, which did not disappear in mutant embryos (data not shown). Both antibodies

showed high accumulation of Smash in the embryonic hindgut, which was also confirmed by RNA

in situ hybridization (Ramrath, 2002), and the denticle belts. However, expressing N-terminally

GFP tagged versions of Smash-PI and Smash-PM in clones in fat body cells, showed clear

localization of GFP-Smash-PI in the nucleus (see Fig.16 B), whereas the large isoform GFP-Smash-

PM showed strong cortical localization and accumulation in a region adjacent to the nuclear

envelope. These localizations could not be shown for endogenous Smash (data not shown). If

there is an in vivo function of Smash in the nucleus, the amount of protein entering the nucleus is

probably below the detection limit. Use of drugs inhibiting nuclear export may lead to

accumulation of endogenous Smash in the nucleus, which would probably be detectable using

anti-Smash antibody. Nuclear localization of Smash would show homology to LMO7 nuclear

function. LMO7 enters the nucleus and regulates expression of muscle-relevant genes, among

DISCUSSION

100

them the LEM domain protein Emerin. Emerin in turn binds to LMO7 inhibiting emerin expression,

indicating a negative feedback mechanism. Mutations in emerin cause Emery-Dreifuss muscular

dystrophy (Nagano et al., 1996; Emery, 2000; Bengtsson and Wilson, 2004). emerin has homologs

in Drosophila as well, the closest one being otefin. Otefin has been shown to be essential for

germline stem cell maintenance (Jiang et al., 2008). Co-IP experiments between Smash and Otefin

could show whether a function in this pathway may be conserved.

4.3 Knockout of smallish

We found that smash gene function is not essential. Homozygous flies are viable and fertile,

although they seem weaker than heterozygous siblings. However, temperature shift to 29°C

increased the lethality score, indicating that stress may influences death rate. So far we have not

tested whether nutritional stress can also change their viability. Differences in the lethality scores

of the two generated alleles smash4.1 and smash35 have not been detected (data not shown). The

latter allele was generated recently, as the gene annotation release 5.40 in 2011 indicated that

the gene annotation was not correct (see 1.5). The smash4.1 allele represents a C-terminal

truncation and therefore cannot be considered a classical null allele. So far we do not know

whether the N-terminal part is expressed in smash4.1. In this regard, by use of the newly produced

anti-Smash N-term antibody we can likely answer this question. Here it would be also interesting

to examine the subcellular localization of the N-terminus. If it is expressed and localized at the

membrane it would clearly show localization independent of Baz.

As the knockout of smash did not lead to lethality, mislocalization of polarity markers was not

expected. Staining for markers such as Baz or DE-Cad showed that cell polarity is intact and AJs

are formed. Examining protein levels did not reveal any change in accumulation of Baz, DE-Cad

and Arm (Fig.21).

The observed lethality scores are comparable with those reported for a knockout of LMO7, where

homozygous mice show up to 40% lethality between birth and weaning. Surviving mice normally

give rise to progeny (Tanaka-Okamoto et al., 2009). In this allele the PDZ domain was precisely

excised, which may disrupt the functionality of the protein. Another LMO7 allele was reported

carrying an 800 kb deletion. Parts of the neighboring gene Uchl3 were taken out as well. Only 40%

of mice were born alive carrying this deletion. Homozygous mice showed growth retardation and

DISCUSSION

101

muscular as well as retinal degeneration. However, latter observation is most likely caused by the

Uchl3 mutation, as single mutants for this gene showed a comparable phenotype (Semenova et

al., 2003). Our smash knockouts do not exhibit obvious defects, but respective electron

microscopy analysis of muscle fibers and the ommatidia has not been performed so far. Subtle

morphological defects are probably not detectable by conventional light and confocal microscopy.

4.4 Overexpression of Smash

We found that in contrast to the smash knockout, overexpression of N-terminally GFP tagged

versions of Smash dramatically increased the lethality score (using tub Gal4 as a strong driver

line). The short isoform, GFP-Smash-PI caused no embryonic lethality but many individuals died

during larval and pupal development. Expression of the larger isoform GFP-Smash-PM led to

embryonic lethality for approximately 50% of individuals, and no hatched larvae developed to

pupation. Since the activity of the UAS/Gal4 system is temperature-dependent, a shift to 18°C

reduces transgene expression. Even at this temperature not a single fly eclosed, emphasizing the

high lethality caused by expression of GFP-Smash-PM. We found that eclosing flies expressing

GFP-Smash-PI were strongly reduced in size (see Fig.23 A). With regards to this, expression of

GFP-Smash-PM by en Gal4 showed that cells overexpressing GFP-Smash-PM were smaller in

comparison to their neighboring non expressing cells. The apical surface area of expressing cells

was significantly decreased. These observations indicate a potential function of smash in apical

constriction, which is a process important for an epithelium undergoing morphogenetic changes

(e.g. invagination, see 1.2). Cuticles of embryos which expressed GFP-Smash-PM under the

control of tub Gal4 showed anterior holes, dorsal holes, or both. It would be interesting to test

whether Smash transgenes lacking the LIM domain or the PDZ binding motif show the same

phenotypes upon overexpression. This could clarify the functional relevance of the C-terminal

part of Smash with regards to the observations.

It would be of big interest whether smash mutant clones would show the opposing effect, i.e. a

widening of the apical surface. However genetic methods are restricted. FRT recombination is the

method of choice for clone induction. Recombination of smash alleles with a FRT is not easy, as

the relevant FRT chromosome used for this chromosome arm is FRT82B. The genomic locus of

smash (3R 82D-E) is very close to this FRT and we were not able to obtain a single recombinant so

far. Another possibility would be the generation of clones with a translocated genomic locus of

DISCUSSION

102

smash (e.g. on the second chromosome) in the smash mutant background. However, only a single

genomic clone is available carrying the entire smash gene locus and generation of respective

transformants did not work so far.

4.5 Smash forms a complex with Src42A in vivo and interacts genetically

with Src64B

In a Drosophila wide yeast two-hybrid screen it was previously shown that the non-receptor

tyrosine kinase Src42A is a potential binding partner of Smash-PI and Smash-PJ respectively (Giot

et al., 2003). Src42A was shown to be expressed in epithelia (Takahashi et al., 2005) and its

activated form, pSrc is restricted to AJs and sites of morphogenetic rearrangements (Shindo et al.,

2008). Due to this we were interested in investigating the possible binding of Smash and Src42A.

We previously showed that Src42A as well as Src64B can bind to Smash-PI in vitro in S2 cells

(Beati, 2009). Furthermore, results presented in this thesis indicate that Src42A phosphorylates

Smash-PI in vitro in cell culture, which is specific for Src42A, as Src64B mediated phosphorylation

was barely detectable (see Fig.30). This finding was strongly supported by the fact that

endogeneous Smash protein showed tyrosine phosphorylation in vivo in embryos as well (see

Fig.30 B). We have not tested whether Smash is still phosphorylated in Src42A mutants. As

described in the introduction SFKs have many overlapping functions. Therefore it would be

interesting to check the tyrosine phosphorylation state of Smash in these respective mutations.

We found that deletion mutants of Src42A can still bind to Smash. As Smash is likely

phosphorylated at three residues, the following mechanism is imaginable: Src42A could bind

Smash via its SH3 domain and thereby phosphorylate Smash on an initial residue. Concomitant

binding to this phosphorylated tyrosine, likely mediated via the SH2 domain could cause intense

phosphorylation of Smash. This would suggest a positive feedback loop and would fit into a

proposed model for phosphotyrosine signaling (Lim and Pawson, 2010). This hypothesis is

supported by the fact that deletion of the SH2 domain showed reduction in binding and

phosphorylation of Smash (see Fig.31 B). However, the developmental relevance of this

interaction is not clear so far.

It was shown that Abelson, another tyrosine kinase, is able to phosphorylate Arm Y667, which is

also phosphorylated by Src42A (Takahashi et al., 2005; Tamada et al., 2012). However, Abelson

DISCUSSION

103

can also induce activation of Src42A, as expression of abelson transgenes with dpp Gal4 in the

wing discs led to specific activation of Src42A in the expression stripe (Singh et al., 2010). We

performed comparable experiments using anti-pSrc antibody (Shindo et al., 2008) but could not

detect changes in pSrc levels upon overexpression of Smash (data not shown). From these results

we conclude that smash is probably not involved in activation of Src42A. smash could also have

functions in inhibiting Src42A activity, as LMO7 was implicated in functioning as a tumor

suppressor. Here 90 week old LMO7 deficient mice showed development of adenocarcinomas in

the lung (Tanaka-Okamoto et al., 2009). To address this possibility we expressed Src and Smash in

the eye by pGMR Gal4. We found contradictory phenotypes: Src42A-HA and Src64B-HA

expression resulted in rough eyes which were slighty enlarged, stronger phenotypes were

observed upon expression of constitutively active forms of Src42A (carrying Y to F mutation,

thereby losing the ability to be phosphorylated by Csk, see 1.3). However, expression of GFP-

Smash-PM resulted in rough eyes as well, which were smaller in size. The same observation was

made upon expression of GFP-Smash-PI. The size decrease was observable most strongly in the

a/p axis of the eye (see Fig.28). However, co-expression of Src42A-HA and GFP-Smash-PM could

reduce the enlarged eye phenotype caused by Src42A-HA expression, which is likely caused by

additive effects. If Smash has the potential to inhibit Src42A function it would be interesting to

test whether the Src42A-HA mediated overexpression phenotype is enhanced in eyes mutant for

smash. Unfortunately we have not performed this experiment yet.

Src42A and Src64B function redundantly in several morphogenetic processes like dorsal closure or

germband retraction (Tateno et al., 2000; Takahashi et al., 2005). Single mutants do not exhibit

strong defects, whereas double mutants do. We generated double mutants of smash with Src42A

and Src64B. Reported mutations in Src42A are lethal (Takahashi et al., 2005) and one copy of

Src42A gene function is still sufficient for survival in the smash mutant background. If both

proteins are acting in the same pathway, this would explain this observation. In contrast, a double

mutant for smash4.1 and Src64BKO resulted in a high lethality score. Only 30% of larvae hatched,

which usually died immediately. However, some eclosing escaper flies were observed. Epithelial

polarization and integrity was not affected. We cannot exclude that the observed lethality is

caused by disruption of cell polarity, as we only analyzed zygotic mutants. Homozygous escaper

flies could not be kept as a stock and did not give rise to any progeny. Maternal supply of either

Src64B or smash is probably important and rescues epithelial polarity in embryos of the first

generation of heterozygous parents. As escaper flies were extremely rare, we could not analyze

embryos in the second generation, lacking maternal supply. Currently we also do not know

DISCUSSION

104

whether the C-terminal truncated allele smash4.1 retains part of its function. Another double

mutant with the full knockout allele smash35 could increase embryonic lethality and possible

defects. As discussed above, we were unable to generate a recombinant for the FRT82B with

smash4.1 for clone induction. As the Src64B gene locus is on the left arm of the 3rd chromosome,

clones mutant for both mutations are not inducible easily.

We tried to rescue the lethality of Src64B smash double mutants by ubiquitous expression of GFP-

Smash-PM under the control of da Gal4, which could not rescue the observed lethality. As the

gene annotation release 5.40 lists 10 isoforms, we cannot exclude isoform specific functions of

smash. Furthermore, overexpression of Smash might have a negative effect on survival, as we

found that strong overexpression led to 50% lethality during embryogenesis (see Fig.22 B). Flies

were viable expressing the N-terminally tagged version of GFP-Smash-PM with da Gal4, although

they were slightly reduced in size (data not shown). However the genomic background could be

sensitized due to the Src64B mutation. A rescue could be performed with a transgene carrying the

genomic smash locus on a different chromosome as well, but as mentioned above we currently

do not have these flies. However, in this way we would circumvent the problem of functions of

different isoforms, and expression levels of smash would better reflect endogenous levels. A

rescue with Src64B transgenes has not been tested yet.

CONCLUSION AND FUTURE PERSPECTIVES

105

5 Conclusion and future perspectives

We showed that smash gene function only plays minor roles during Drosophila development and

is not important for survival of the fly. We generated two different mutant alleles, one

representing a classical null allele due to deletion of the genomic locus of smash and a second

allele truncated at the C-terminus by deletion of the 3’ genomic region. Both mutant alleles do

not result in lethality or obvious epithelial defects. However, it cannot be ruled out that minor

defects have been overlooked. Since we showed that Src42A phosphorylates Smash in vitro, and

Smash is tyrosine phosphorylated in vivo as well, it would be helpful to identify the missing

phosphorylation target sites by mass spectrometry. By our approach we were able to identify Y64

and Y162 as potential phosphorylation sites of the short isoform Smash-PI. Antibodies raised

against phosphorylated Smash could shed light on its subcellular site of function, although we do

not know whether phosphorylation correlates with activation of Smash. In this regard it would be

very interesting to test whether transgenic flies encoding for the smash gene locus have the

ability to rescue smash Src64B double mutants. A respective transgene mutated in the

phosphorylation sites would clearly show the importance of the Smash Src42A interaction and its

phosphorylation. In the context of the interaction with Src42A it was shown that the vertebrate

homolog of Baz, Par3, is phosphorylated by c-Src, which is the closest homolog of Src42A. Par3

tyrosine phosphorylation results in the dissociation of LIM kinase 2, which in turn regulates

phosphorylation of cofilin and thereby delays TJ assembly (Wang et al., 2006). We could not find a

link between Smash function and Baz in this regard. Baz co-immunoprecipitates with Src42A as

well and showed tyrosine phosphorylation in vitro in cell culture, likely independent of Smash

(data not shown). This observation was made after co-expressing Baz and Src42A in S2 cells and

analyzed in a similar manner to the Smash and Src42A interaction. Since we did not downregulate

Smash protein levels in this system we cannot exclude a function for Smash in the complex

formation of Baz and Src42A. So far we have not tested whether Baz can associate with Src42A in

the absence of Smash. The easiest way would be to test Baz association with Src42A, as well as

phosphorylation of Baz in vivo in embryos compared to smash mutants.

Contradictory to loss of smash, overexpression resulted in a dramatic increase in embryonic

lethality. Overexpression of GFP-Smash-PM using en Gal4 as a driver line resulted in cells that are

smaller and decreased in their apical surface area, which likely represents an apical constriction


106

phenotype. Apical constriction is a process that depends on the Actin/Myosin network, which lies

beneath the AJs (see 1.2). This process is of importance for morphogenetic rearrangements, as

contraction exerts force on AJs and their relocation results in cell shape changes. Apical

constriction is regulated by Rho-associated kinase (ROCK) activity, which is involved in the

phosphorylation of Myosin light chain, thereby enhancing the Actin/Myosin contractility (Riento

and Ridley, 2003; Vicente-Manzanares et al., 2009). For example, expression of Spaghetti-squash

the Drosophila homolog of Myosin light chain, mutated in its phosphorylation sites results in

apical expansion if mutated to non phosphorylatable alanine, whereas glutamate exchange

results in apical constriction (Zimmerman et al., 2010). Recently it was shown that aPKC is

negatively involved in apical constriction, as it is recruited to the apical membrane not only by

Par3 but also by Willin. Simultaneous depletion of Willin and Par3 resulted in loss of aPKC at the

apical membrane and induced apical constriction. aPKC had been shown to phosphorylate ROCK,

thereby reducing the junctional localization of ROCK explaining these findings (Ishiuchi and

Takeichi, 2011). If the observed phenotype caused by Smash represents apical constriction it is

still elusive how Smash might function in this pathway. So far we have not focused on potential

interacting proteins of the Actin/Myosin network. Candidates would be Canoe, the Drosophila

homolog of Afadin and α-Actinin. Both proteins were shown to bind LMO7 (Ooshio et al., 2004).

Furthermore, Canoe supports a link between the Actin cytoskeleton and AJs during apical

constriction and mutants for canoe show dorsal closure defects (Takahashi et al., 1998; Sawyer et

al., 2009). Whether Src42A plays a part in this pathway as well could be tested by expressing

Smash-PM mutated at its phosphorylation sites. Loss of the smaller cell phenotype would clearly

place Src42A in this context.

The classical model of cell-cell adhesion was thought to be mediated by a stable connection

between AJs and the cytoskeleton through α-Cat (see 1.2). However, it was shown that a

quaternary complex of E-Cad-β-Cat-α-Cat-Actin does not exist and that the link between AJs and

the cytoskeleton is likely mediated by several different interacting modules (e.g. Afadin) resulting

in a highly dynamic connection (Drees et al., 2005; Yamada et al., 2005). The cortical Actin based

cytoskeleton was also reported to be more dynamic than the AJs by FRAP (fluorescene recovery

after photobleaching) (Gates and Peifer, 2005; Yamada et al., 2005). Smash might represent a so

far undescribed link between AJs and the cytoskeleton. It would be interesting to perform FRAP

analysis of the cortical Actin cytoskeleton upon overexpression of Smash and record its activity.


107

We provide evidence that smash gene function is not essential but cannot exclude whether smash

functions redundantly with other genes. The vertebrate homolog LMO7 was duplicated and

shows a paralog LIMCH1 (Friedberg, 2009, 2010). It is possible that a double knockout of LMO7

and LIMCH1 results in lethality and severe defects, if both genes function redundantly. However, a

LIMCH1 mutant is not reported. It would be important to search for similar proteins encoded by

the Drosophila genome, although smash appears not to be duplicated. Generation of different

double mutants with smash might exhibit specific defects. However, blast search does not show

genes similar to smash. In this context proteins with related domain composition would be of

interest as well.

CVIII

6 References

Atwood, S.X., Chabu, C., Penkert, R.R., Doe, C.Q., and Prehoda, K.E. (2007). Cdc42 acts downstream of Bazooka to regulate neuroblast polarity through Par-6 aPKC. Journal of Cell Science 120, 3200–3206.

Bach, I. (2000). The LIM domain: regulation by association. Mechanisms of Development 91, 5–17.

Bachmann, A. and Knust, E. (2008). The use of P-element transposons to generate transgenic flies. Methods in Molecular Biology 420, 61-77.

Beati, H. (2009). Die Rolle der Bazooka-CG31534 Interaktion bei der Etablierung der Zellpolarität in Drosophila melanogaster. Diplomarbeit an der Georg-August-Universität Göttingen, Göttingen. Behrens, J., Vakaet, L., Friis, R., Winterhager, E., Van Roy, F., Mareel, M.M., and Birchmeier, W. (1993). Loss of epithelial differentiation and gain of invasiveness correlates with tyrosine phosphorylation of the E-cadherin/beta-catenin complex in cells transformed with a temperature-sensitive v-SRC gene. The Journal of Cell Biology 120, 757–766.

Bengtsson, L., and Wilson, K.L. (2004). Multiple and surprising new functions for emerin, a nuclear membrane protein. Current Opinion in Cell Biology 16, 73–79.

Benton, R., and Johnston, D.S. (2003). A conserved oligomerization domain in drosophila Bazooka/PAR-3 is important for apical localization and epithelial polarity. Current Biology 13, 1330–1334.

Betschinger, J., Mechtler, K., and Knoblich, J.A. (2003). The Par complex directs asymmetric cell division by phosphorylating the cytoskeletal protein Lgl. Nature 422, 326–330.

Bilder, D., Li, M., and Perrimon, N. (2000). Cooperative regulation of cell polarity and growth by Drosophila tumor suppressors. Science 289, 113–116.

Bilder, D., and Perrimon, N. (2000). Localization of apical epithelial determinants by the basolateral PDZ protein Scribble. Nature 403, 676–680.

Bilder, D., Schober, M., and Perrimon, N. (2003). Integrated activity of PDZ protein complexes regulates epithelial polarity. Nature Cell Biology 5, 53–58.

Bloor, J.W., and Brown, N.H. (1998). Genetic analysis of the Drosophila alphaPS2 integrin subunit reveals discrete adhesive, morphogenetic and sarcomeric functions. Genetics 148, 1127–1142.

Bodenmiller, B., Malmstrom, J., Gerrits, B., Campbell, D., Lam, H., Schmidt, A., Rinner, O., Mueller, L.N., Shannon, P.T., Pedrioli, P.G., et al. (2007). PhosphoPep--a phosphoproteome resource for systems biology research in Drosophila Kc167 cells. Molecular Systems Biology 3.

Boschek, C.B., Jockusch, B.M., Friis, R.R., Back, R., Grundmann, E., and Bauer, H. (1981). Early changes in the distribution and organization of microfilament proteins during cell transformation. Cell 24, 175–184.

CIX

Boyer, B., Roche, S., Denoyelle, M., and Thiery, J.P. (1997). Src and Ras are involved in separate pathways in epithelial cell scattering. The EMBO Journal 16, 5904–5913.

Brand, A.H., and Perrimon, N. (1993). Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118, 401–415.

Brown, M.T., and Cooper, J.A. (1996). Regulation, substrates and functions of src. Biochimica Et Biophysica Acta. 1287, 121–149.

Cenciarelli, C., Chiaur, D.S., Guardavaccaro, D., Parks, W., Vidal, M., and Pagano, M. (1999). Identification of a family of human F-box proteins. Current Biology 9, 1177–1179.

Chen, J., Shi, X., Padmanabhan, R., Wang, Q., Wu, Z., Stevenson, S.C., Hild, M., Garza, D., and Li, H. (2008). Identification of novel modulators of mitochondrial function by a genome-wide RNAi screen in Drosophila melanogaster. Genome Research 18, 123–136.

Cherret, C., Furutani-Seiki, M., and Bagby, S. (2012). The Hippo pathway: key interaction and catalytic domains in organ growth control, stem cell self-renewal and tissue regeneration. Essays in Biochemistry 53, 111-127.

Chihara, T., and Hayashi, S. (2000). Control of tracheal tubulogenesis by Wingless signaling. Development 127, 4433–4442.

Cizkova, M., Cizeron-Clairac, G., Vacher, S., Susini, A., Andrieu, C., Lidereau, R., and Bièche, I. (2010). Gene expression profiling reveals new aspects of PIK3CA mutation in ERalpha-positive breast cancer: major implication of the Wnt signaling pathway. PLoS Genetics 5, e15647.

Dawes-Hoang, R.E., Parmar, K.M., Christiansen, A.E., Phelps, C.B., Brand, A.H., and Wieschaus, E.F. (2005). folded gastrulation, cell shape change and the control of myosin localization. Development 132, 4165–4178.

Dedeic, Z., Cetera, M., Cohen, T. V, and Holaska, J.M. (2011). Emerin inhibits Lmo7 binding to the Pax3 and MyoD promoters and expression of myoblast proliferation genes. Journal of Cell Science 124, 1691–1702.

Desprat, N., Supatto, W., Pouille, P.-A., Beaurepaire, E., and Farge, E. (2008). Tissue deformation modulates twist expression to determine anterior midgut differentiation in Drosophila embryos. Developmental Cell 15, 470–477.

Dodson, G.S., Guarnieri, D.J., and Simon, M.A. (1998). Src64 is required for ovarian ring canal morphogenesis during Drosophila oogenesis. Development 125, 2883–2892.

Doe, C.Q. (2008). Neural stem cells: balancing self-renewal with differentiation. Development 135, 1575–1587.

Drees, F., Pokutta, S., Yamada, S., Nelson, W.J., and Weis, W.I. (2005). Alpha-catenin is a molecular switch that binds E-Cadherin-beta-catenin and regulates actin-filament assembly. Cell 123, 903–915.

CX

Emery, A.E.H. (2000). Emery-Dreifuss muscular dystrophy - a 40 year retrospective. Neuromuscular Disorders 10, 228–232.

Engen, J.R., Wales, T.E., Hochrein, J.M., Meyn, M.A. 3rd, Banu Ozkan, S., Bahar, I., and Smithgall, T.E. (2008). Structure and dynamic regulation of Src-family kinases. Cellular and Molecular Life Sciences 65, 3058–3073.

Fanning, A.S., and Anderson, J.M. (1999). PDZ domains: fundamental building blocks in the organization of protein complexes at the plasma membrane. The Journal of Clinical Investigation 103, 767–772.

Freyd, G., Kim, S.K., and Horvitz, H.R. (1990). Novel cysteine-rich motif and homeodomain in the product of the Caenorhabditis elegans cell lineage gene lin-11. Nature 344, 876–879.

Friedberg, F. (2009). Alternative splicing for members of human mosaic domain superfamilies. I. The CH and LIM domains containing group of proteins. Molecular Biology Reports 36, 1059–1081.

Friedberg, F. (2010). Singlet CH domain containing human multidomain proteins: an inventory. Molecular Biology Reports 37, 1531–1539.

Förster, D., and Luschnig, S. (2012). Src42A-dependent polarized cell shape changes mediate epithelial tube elongation in Drosophila. Nature Cell Biology 14, 526–534.

Gates, J., and Peifer, M. (2005). Can 1000 reviews be wrong? Actin, alpha-Catenin, and adherens junctions. Cell 123, 769–772.

Giot, L., Bader, J.S., Brouwer, C., Chaudhuri, A., Kuang, B., Li, Y., Hao, L., Ooi, C.E., Godwin, B., Vitols, E., et al. (2003). A protein interaction map of Drosophila melanogaster. Science 302, 1727–1736.

Groth, A.C., Fish, M., Nusse, R., and Calos, M.P. (2004). Construction of transgenic Drosophila by using the site-specific integrase from phage phiC31. Genetics 166, 1775–1782.

Hamaguchi, M., Matsuyoshi, N., Ohnishi, Y., Gotoh, B., Takeichi, M., and Nagai, Y. (1993). p60v-src causes tyrosine phosphorylation and inactivation of the N-cadherin-catenin cell adhesion system. The EMBO Journal 12, 307–314.

Harris, B.Z., and Lim, W.A. (2001). Mechanism and role of PDZ domains in signaling complex assembly. Journal of Cell Science 114, 3219–3231.

Harris, T.J.C., and Peifer, M. (2005). The positioning and segregation of apical cues during epithelial polarity establishment in Drosophila. The Journal of Cell Biology 170, 813–823.

Holaska, J.M., Rais-Bahrami, S., and Wilson, K.L. (2006). Lmo7 is an emerin-binding protein that regulates the transcription of emerin and many other muscle-relevant genes. Human Molecular Genetics 15, 3459–3472.

Holaska, J.M., and Wilson, K.L. (2006). Multiple roles for emerin: implications for Emery-Dreifuss muscular dystrophy. The Anatomical Record. Part A, Discoveries in Molecular, Cellular, and Evolutionary Biology. 288, 676–680.

CXI

Horikoshi, Y., Suzuki, A., Yamanaka, T., Sasaki, K., Mizuno, K., Sawada, H., Yonemura, S., and Ohno, S. (2009). Interaction between PAR-3 and the aPKC-PAR-6 complex is indispensable for apical domain development of epithelial cells. Journal of Cell Science 122, 1595–1606.

Huang, J., Zhou, W., Dong, W., Watson, A.M., and Hong, Y. (2009). Directed , efficient , and versatile modifications of the Drosophila genome by genomic engineering. Proceedings of the National Academy of Sciences of the United States of America 106, 8284–8289.

Hurov, J.B., Watkins, J.L., and Piwnica-Worms, H. (2004). Atypical PKC phosphorylates PAR-1 kinases to regulate localization and activity. Current Biology 14, 736–741.

Ia, K.K., Mills, R.D., Hossain, M., Chan, K.C., Jarasrassamee, B., Jorissen, R.N., and Cheng, H.C. (2010). Structural elements and allosteric mechanisms governing regulation and catalysis of CSK-family kinases and their inhibition of Src-family kinases. Growth Factors 28, 329-350.

Ishiuchi, T., and Takeichi, M. (2011). Willin and Par3 cooperatively regulate epithelial apical constriction through aPKC-mediated ROCK phosphorylation. Nature Cell Biology 13, 860–866.

Jacinto, A., Woolner, S., and Martin, P. (2002). Dynamic analysis of dorsal closure in Drosophila: from genetics to cell biology. Developmental Cell 3, 9–19.

Jeleń, F., Oleksy, A., Śmietana, K., and Otlewski, J. (2003). PDZ domains — common players in the cell signaling. Acta Biochimica Polonica 50, 985–1017.

Jiang, X., Xia, L., Chen, D., Yang, Y., Huang, H., Yang, L., Zhao, Q., Shen, L., Wang, J., and Chen, D. (2008). Otefin , a nuclear membrane protein, determines the fate of germline stem cells in Drosophila via interaction with Smad complexes. Developmental Cell 14, 494–506.

Johnson, K., and Wodarz, A. (2003). A genetic hierarchy controlling cell polarity. Nature Cell Biology 5, 12–14.

Juarez, M.T., Patterson, R.A., Sandoval-Guillen, E., and McGinnis, W. (2011). Duox, Flotillin-2, and Src42A are required to activate or delimit the spread of the transcriptional response to epidermal wounds in Drosophila. PLoS Genetics 7, e1002424.

Kadrmas, J.L., and Beckerle, M.C. (2004). The LIM domain: from the cytoskeleton to the nucleus. Nature Reviews. Molecular Cell Biology. 5, 920–931.

Karlsson, O., Thor, S., Norberg, T., Ohlsson, H., and Edlund, T. (1990). Insulin gene enhancer binding protein Isl-1 is a member of a novel class of proteins containing both a homeo- and a Cys-His domain. Nature 344, 879–882.

Khurana, T., Khurana, B., and Noegel, A.A. (2002). LIM proteins: association with the actin cytoskeleton. Protoplasma 219, 1–12.

Kim, L., and Wong, T.W. (1998). Growth factor-dependent phosphorylation of the actin-binding protein cortactin is mediated by the cytoplasmic tyrosine kinase FER. The Journal of Biological Chemistry 273, 23542–23548.

CXII

Knust, E., and Bossinger, O. (2002). Composition and formation of intercellular junctions in epithelial cells. Science 298, 1955–1959.

Krahn, M.P., Bückers, J., Kastrup, L., and Wodarz, A. (2010a). Formation of a Bazooka-Stardust complex is essential for plasma membrane polarity in epithelia. The Journal of Cell Biology 190, 751–760.

Krahn, M.P., Klopfenstein, D.R., Fischer, N., and Wodarz, A. (2010b). Membrane targeting of Bazooka/PAR-3 is mediated by direct binding to phosphoinositide lipids. Current Biology 20, 636–642.

Kusakabe, M., and Nishida, E. (2004). The polarity-inducing kinase Par-1 controls Xenopus gastrulation in cooperation with 14-3-3 and aPKC. The EMBO Journal 23, 4190–4201.

Laplante, C., and Nilson, L.A. (2011). Asymmetric distribution of Echinoid defines the epidermal leading edge during Drosophila dorsal closure. The Journal of Cell Biology 192, 335–348.

Laprise, P., Paul, S.M., Boulanger, J., Robbins, R.M., Beitel, G.J., and Tepass, U. (2010). Epithelial polarity proteins regulate Drosophila tracheal tube size in parallel to the luminal matrix pathway. Current Biology : CB 20, 55–61.

Lilien, J., Balsamo, J., Arregui, C., and Xu, G. (2002). Turn-off, drop-out: functional state switching of cadherins. Developmental Dynamics 224, 18–29.

Lim, W.A., and Pawson, T. (2010). Phosphotyrosine signaling: evolving a new cellular communication system. Cell 142, 661–667.

Llimargas, M. (2000). Wingless and its signalling pathway have common and separable functions during tracheal development. Development 127, 4407–4417.

Lu, X., and Li, Y. (1999). Drosophila Src42A is a negative regulator of RTK signaling. Developmental Biology 208, 233–243.

McCaffrey, L.M., and Macara, I.G. (2009). The Par3/aPKC interaction is essential for end bud remodeling and progenitor differentiation during mammary gland morphogenesis. Genes & Development 23, 1450–1460.

McGill, M.A., McKinley, R.F.A., and Harris, T.J.C. (2009). Independent cadherin-catenin and Bazooka clusters interact to assemble adherens junctions. The Journal of Cell Biology 185, 787–796.

Morais-de-Sá, E., Mirouse, V., and St Johnston, D. (2010). aPKC phosphorylation of Bazooka defines the apical/lateral border in Drosophila epithelial cells. Cell 141, 509–523.

Murray, M.J., Davidson, C.M., Hayward, N.M., and Brand, A.H. (2006). The Fes/Fer non-receptor tyrosine kinase cooperates with Src42A to regulate dorsal closure in Drosophila. Development 133, 3063–3073.

Müller, H.-A.J. (2000). Genetic control of epithelial cell polarity: lessons from Drosophila. Developmental Dynamics 218, 52–67.

CXIII

Müller, H.-A.J., and Wieschaus, E. (1996). armadillo, bazooka, and stardust are critical for early stages in formation of the zonula adherens and maintenance of the polarized blastoderm epithelium in Drosophila. The Journal of Cell Biology 134, 149–163.

Nagano, A., Koga, R., Ogawa, M., Kurano, Y., Kawada, J., Okada, R., Hayashi, Y.K., Tsukahara, T., and Arahata, K. (1996). Emerin deficiency at the nuclear membrane in patients with Emery-Dreifuss muscular dystrophy. Nature Genetics 12, 254–259.

Nakamura, H., Hori, K., Tanaka-Okamoto, M., Higashiyama, M., Itoh, Y., Inoue, M., Morinaka, S., and Miyoshi, J. (2011). Decreased expression of LMO7 and its clinicopathological significance in human lung adenocarcinoma. Experimental And Therapeutic Medicine 2, 1053–1057.

Nakamura, H., Mukai, M., Komatsu, K., Tanaka-Okamoto, M., Itoh, Y., Ishizaki, H., Tatsuta, M., Inoue, M., and Miyoshi, J. (2005). Transforming growth factor-β1 induces LMO7 while enhancing the invasiveness of rat ascites hepatoma cells. Cancer Letters 220, 95–99.

Nam, S.-C., and Choi, K.-W. (2003). Interaction of Par-6 and Crumbs complexes is essential for photoreceptor morphogenesis in Drosophila. Development 130, 4363–4372.

Nelson, K.S., Khan, Z., Molnár, I., Mihály, J., Kaschube, M., and Beitel, G.J. (2012). Drosophila Src regulates anisotropic apical surface growth to control epithelial tube size. Nature Cell Biology 14, 518–525.

Neugebauer, J. (2007). Untersuchungen zur Funktion des Gens CG31534 bei der Entwicklung ektodermaler Epithelien in Drosophila. Diplomarbeit an der Georg-August-Universität Göttingen, Göttingen. Ochoa-Espinosa, A., Baer, M.M., and Affolter, M. (2012). Tubulogenesis: Src42A goes to great lengths in tube elongation. Current Biology 22, R446–449.

Ooshio, T., Irie, K., Morimoto, K., Fukuhara, A., Imai, T., and Takai, Y. (2004). Involvement of LMO7 in the association of two cell-cell adhesion molecules, nectin and E-cadherin, through afadin and alpha-actinin in epithelial cells. The Journal of Biological Chemistry 279, 31365–31373.

Ott, E.B., Van den Akker, N.M.S., Sakalis, P.A., Gittenberger-de Groot, A.C., Te Velthuis, A.J.W., and Bagowski, C.P. (2008). The lim domain only protein 7 is important in zebrafish heart development. Developmental Dynamics 237, 3940–3952.

O’Reilly, A.M., Ballew, A.C., Miyazawa, B., Stocker, H., Hafen, E., and Simon, M.A. (2006). Csk differentially regulates Src64 during distinct morphological events in Drosophila germ cells. Development 133, 2627–2638.

Parsons, S.J., and Parsons, J.T. (2004). Src family kinases , key regulators of signal transduction. Oncogene 23, 7906–7909.

Pedraza, L.G., Stewart, R.A., Li, D.-M., and Xu, T. (2004). Drosophila Src-family kinases function with Csk to regulate cell proliferation and apoptosis. Oncogene 23, 4754–4762.

Petronczki, M., and Knoblich, J.A. (2001). DmPAR-6 directs epithelial polarity and asymmetric cell division of neuroblasts in Drosophila. Nature Cell Biology 3, 43–49.

CXIV

Piedra, J., Miravet, S., Castaño, J., Pálmer, H.G., Heisterkamp, N., Herreros, A.G. de, and Duñach, M. (2003). p120 Catenin-associated Fer and Fyn tyrosine kinases regulate beta-catenin Tyr-142 phosphorylation and beta-catenin-alpha-catenin interaction. Molecular and Cellular Biology 23, 2287–2297.

Pingoud, A., and Jeltsch, A. (2001). Structure and function of type II restriction endonucleases. Nucleic Acids Research 29, 3705–3727.

Plant, P.J., Fawcett, J.P., Lin, D.C.C., Holdorf, A.D., Binns, K., Kulkarni, S., and Pawson, T. (2003). A polarity complex of mPar-6 and atypical PKC binds, phosphorylates and regulates mammalian Lgl. Nature Cell Biology 5, 301–308.

Ramrath, A. (2002). Isolierung und Charakterisierung von Bindungspartnern des PDZDomänen- Proteins BAZOOKA aus Drosophila melanogaster. Inaugural-Dissertation in Mathematisch-Naturwissenschaftlicher Fakultät der Heinrich-Heine-Universität Düsseldorf, Düsseldorf. Riento, K., and Ridley, A.J. (2003). Rocks : multifunctional kinases in cell behaviour. Nature Reviews. Molecular Cell Biology. 4, 446–456.

Rozenblum, E., Vahteristo, P., Sandberg, T., Bergthorsson, J.T., Syrjakoski, K., Weaver, D., Haraldsson, K., Johannsdottir, H.K., Vehmanen, P., Nigam, S., et al. (2002). A genomic map of a 6-Mb region at 13q21-q22 implicated in cancer development: identification and characterization of candidate genes. Human Genetics 110, 111–121.

Sasaki, M., Tsuji, N., Furuya, M., Kondoh, K., Kamagata, C., Kobayashi, D., Yagihashi, A., and Watanabe, N. (2003). PCD1, a novel gene containing PDZ and LIM domains, is overexpressed in human breast cancer and linked to lymph node metastasis. Anticancer Research 23, 2717-2721.

Sawyer, J.K., Harris, N.J., Slep, K.C., Gaul, U., and Peifer, M. (2009). The Drosophila afadin homologue Canoe regulates linkage of the actin cytoskeleton to adherens junctions during apical constriction. The Journal of Cell Biology 186, 57–73.

Semenova, E., Wang, X., Jablonski, M.M., Levorse, J., and Tilghman, S.M. (2003). An engineered 800 kilobase deletion of Uchl3 and Lmo7 on mouse chromosome 14 causes defects in viability, postnatal growth and degeneration of muscle and retina. Human Molecular Genetics 12, 1301–1312.

Sewell, W., Sparrow, D.B., Smith, A.J., Gonzalez, D.M., Rappaport, E.F., Dunwoodie, S.L., and Kusumi, K. (2010). Cyclical expression of the Notch/Wnt regulator Nrarp requires modulation by Dll3 in somitogenesis. Developmental Biology 329, 400–409.

Sheng, M., and Sala, C. (2001). PDZ domains and the organization of supramolecular complexes. Annual Review of Neuroscience 24, 1–29.

Shindo, M., Wada, H., Kaido, M., Tateno, M., Aigaki, T., Tsuda, L., and Hayashi, S. (2008). Dual function of Src in the maintenance of adherens junctions during tracheal epithelial morphogenesis. Development 135, 1355–1364.

Simon, M.A., Drees, B., Kornberg, T., and Bishop, J.M. (1985). The nucleotide sequence and the tissue-specific expression of Drosophila c-src. Cell 42, 831–840.

CXV

Singh, J., Aaronson, S.A., and Mlodzik, M. (2010). Drosophila Abelson kinase mediates cell invasion and proliferation through two distinct MAPK pathways. Oncogene 29, 4033–4045.

Sotillos, S., Díaz-Meco, M.T., Caminero, E., Moscat, J., and Campuzano, S. (2004). DaPKC-dependent phosphorylation of Crumbs is required for epithelial cell polarity in Drosophila. The Journal of Cell Biology 166, 549–557.

Von Stein, W., Ramrath, A., Grimm, A., Müller-Borg, M., and Wodarz, A. (2005). Direct association of Bazooka/PAR-3 with the lipid phosphatase PTEN reveals a link between the PAR/aPKC complex and phosphoinositide signaling. Development 132, 1675–1686.

Strong, T.C., and Thomas, J.H. (2011). Maternal and zygotic requirements for src64 during Drosophila cellularization. Genesis 49, 912–918.

Suzuki, A., Hirata, M., Kamimura, K., Maniwa, R., Yamanaka, T., Mizuno, K., Kishikawa, M., Hirose, H., Amano, Y., Izumi, N., et al. (2004). aPKC acts upstream of PAR-1b in both the establishment and maintenance of mammalian epithelial polarity. Current Biology 14, 1425–1435.

Suzuki, A., and Ohno, S. (2006). The PAR-aPKC system: lessons in polarity. Journal of Cell Science 119, 979–987.

Takahashi, F., Endo, S., Kojima, T., and Saigo, K. (1996). Regulation of cell-cell contacts in developing Drosophila eyes by Dsrc41, a new, close relative of vertebrate c-src. Genes & Development 10, 1645–1656.

Takahashi, K., Matsuo, T., Katsube, T., Ueda, R., and Yamamoto, D. (1998). Direct binding between two PDZ domain proteins Canoe and ZO-1 and their roles in regulation of the Jun N-terminal kinase pathway in Drosophila morphogenesis. Mechanisms of Development 78, 97–111.

Takahashi, M., Takahashi, F., Ui-Tei, K., Kojima, T., and Saigo, K. (2005). Requirements of genetic interactions between Src42A, armadillo and shotgun, a gene encoding E-cadherin, for normal development in Drosophila. Development 132, 2547–2559.

Takeda, H., Nagafuchi, A., Yonemura, S., Tsukita, S., Behrens, J., Birchmeier, W., and Tsukita, S. (1995). V-src kinase shifts the cadherin-based cell adhesion from the strong to the weak state and beta catenin is not required for the shift. The Journal of Cell Biology 131, 1839–1847.

Tamada, M., Farrell, D.L., and Zallen, J.A. (2012). Abl regulates planar polarized junctional dynamics through β-catenin tyrosine phosphorylation. Developmental Cell 22, 309–319.

Tanaka-Okamoto, M., Hori, K., Ishizaki, H., Hosoi, A., Itoh, Y., Wei, M., Wanibuchi, H., Mizoguchi, A., Nakamura, H., and Miyoshi, J. (2009). Increased susceptibility to spontaneous lung cancer in mice lacking LIM-domain only 7. Cancer Science 100, 608–616.

Tanentzapf, G., and Tepass, U. (2003). Interactions between the crumbs, lethal giant larvae and bazooka pathways in epithelial polarization. Nature Cell Biology 5, 46–52.

Tateno, M., Nishida, Y., and Adachi-Yamada, T. (2000). Regulation of JNK by Src during Drosophila development. Science 287, 324–327.

CXVI

Tepass, U. (1996). Crumbs, a component of the apical membrane, is required for zonula adherens formation in primary epithelia of Drosophila. Developmental Biology 177, 217–225.

Tepass, U. (2012). The apical polarity protein network in Drosophila epithelial cells: regulation of polarity, junctions, morphogenesis, cell growth, and survival. Annual Review of Cell and Developmental Biology 28, 655–685.

Thibault, S.T., Singer, M.A., Miyazaki, W.Y., Milash, B., Dompe, N.A., Singh, C.M., Buchholz, R., Demsky, M., Fawcett, R., Francis-Lang, H.L., et al. (2004). A complementary transposon tool kit for Drosophila melanogaster using P and piggyBac. Nature Genetics 36, 283–287.

Thomas, J.H., and Wieschaus, E. (2004). src64 and tec29 are required for microfilament contraction during Drosophila cellularization. Development 131, 863–871.

Thomas, S.M., and Brugge, J.S. (1997). Cellular functions regulated by Src family kinases. Annual Review of Cell and Developmental Biology 13, 513–609.

Tikhmyanova, N., Tulin, A. V, Roegiers, F., and Golemis, E.A. (2010). Dcas supports cell polarization and cell-cell adhesion complexes in development. PloS One 5, e12369.

Tsai, P.-I., Kao, H.-H., Grabbe, C., Lee, Y.-T., Ghose, A., Lai, T.-T., Peng, K.-P., Van Vactor, D., Palmer, R.H., Chen, R.-H., et al. (2008). Fak56 functions downstream of integrin alphaPS3betanu and suppresses MAPK activation in neuromuscular junction growth. Neural Development 3.

VanHook, A., and Letsou, A. (2008). Head involution in Drosophila: genetic and morphogenetic connections to dorsal closure. Developmental Dynamics 237, 28–38.

Te Velthuis, A.J.W., and Bagowski, C.P. (2007). PDZ and LIM domain-encoding genes: molecular interactions and their role in development. TheScientificWorldJournal 7, 1470–1492.

Te Velthuis, A.J.W., Isogai, T., Gerrits, L., and Bagowski, C.P. (2007). Insights into the molecular evolution of the PDZ/LIM family and identification of a novel conserved protein motif. PloS One 2, e189.

Vicente-Manzanares, M., Ma, X., Adelstein, R.S., and Horwitz, A.R. (2009). Non-muscle myosin II takes centre stage in cell adhesion and migration. Nature Reviews. Molecular Cell Biology. 10, 778–790.

Walther, R.F., and Pichaud, F. (2010). Crumbs/DaPKC-dependent apical exclusion of Bazooka promotes photoreceptor polarity remodeling. Current Biology 20, 1065–1074.

Wang, Y., Du, D., Fang, L., Yang, G., Zhang, C., Zeng, R., Ullrich, A., Lottspeich, F., and Chen, Z. (2006). Tyrosine phosphorylated Par3 regulates epithelial tight junction assembly promoted by EGFR signaling. The EMBO Journal 25, 5058–5070.

Way, J.C., and Chalfie, M. (1988). mec-3, a homeobox-containing gene that specifies differentiation of the touch receptor neurons in C.elegans. Cell 54, 5–16.

CXVII

Wei, S.-Y., Escudero, L.M., Yu, F., Chang, L.-H., Chen, L.-Y., Ho, Y.-H., Lin, C.-M., Chou, C.-S., Chia, W., Modolell, J., et al. (2005). Echinoid is a component of adherens junctions that cooperates with DE-Cadherin to mediate cell adhesion. Developmental Cell 8, 493–504.

Wimmer, E.A. (2003). Innovations: applications of insect transgenesis. Nature Reviews. Genetics. 4, 225–232.

Wodarz, A. (2005). Molecular control of cell polarity and asymmetric cell division in Drosophila neuroblasts. Current Opinion in Cell Biology 17, 475–481.

Wodarz, A. (2008). Extraction and immunoblotting of proteins from embryos. Methods in Molecular Biology 420, 335-345.

Wodarz, A., Grawe, F., and Knust, E. (1993). CRUMBS is involved in the control of apical protein targeting during Drosophila epithelial development. Mechanisms of Development 44, 175–187.

Wodarz, A., Hinz, U., Engelbert, M., and Knust, E. (1995). Expression of crumbs confers apical character on plasma membrane domains of ectodermal epithelia of Drosophila. Cell 82, 67–76.

Wodarz, A., and Huttner, W.B. (2003). Asymmetric cell division during neurogenesis in Drosophila and vertebrates. Mechanisms of Development 120, 1297–1309.

Wodarz, A., Ramrath, A., Grimm, A., and Knust, E. (2000). Drosophila atypical protein kinase C associates with Bazooka and controls polarity of epithelia and neuroblasts. The Journal of Cell Biology 150, 1361–1374.

Wodarz, A., Ramrath, A., Kuchinke, U., and Knust, E. (1999). Bazooka provides an apical cue for Inscuteable localization in Drosophila neuroblasts. Nature 402, 544–547.

Wouda, R.R., Bansraj, M.R.K.S., De Jong, A.W.M., Noordermeer, J.N., and Fradkin, L.G. (2008). Src family kinases are required for WNT5 signaling through the Derailed/RYK receptor in the Drosophila embryonic central nervous system. Development 135, 2277–2287.

Wu, H., and Parsons, J.T. (1993). Cortactin, an 80/85-kilodalton pp60src substrate, is a filamentous actin-binding protein enriched in the cell cortex. The Journal of Cell Biology 120, 1417–1426.

Yamada, S., Pokutta, S., Drees, F., Weis, W.I., and Nelson, W.J. (2005). Deconstructing the cadherin-catenin-actin complex. Cell 123, 889–901.

Yamanaka, T., Horikoshi, Y., Sugiyama, Y., Ishiyama, C., Suzuki, A., Hirose, T., Iwamatsu, A., Shinohara, A., and Ohno, S. (2003). Mammalian Lgl forms a protein complex with PAR-6 and aPKC independently of PAR-3 to regulate epithelial cell polarity. Current Biology 13, 734–743.

Yue, T., Tian, A., and Jiang, J. (2012). The cell adhesion molecule echinoid functions as a tumor suppressor and upstream regulator of the Hippo signaling pathway. Developmental Cell 22, 255–267.

Ziegenfuss, J.S., Biswas, R., Avery, M.A., Hong, K., Sheehan, A.E., Yeung, Y.-G., Stanley, E.R., and Freeman, M.R. (2008). Draper-dependent glial phagocytic activity is mediated by Src and Syk family kinase signalling. Nature 453, 935–939.

CXVIII

Zimmerman, S.G., Thorpe, L.M., Medrano, V.R., Mallozzi, C.A., and McCartney, B.M. (2010). Apical constriction and invagination downstream of the canonical Wnt signaling pathway require Rho1 and Myosin II. Developmental Biology 340, 54–66.

functional analysis of the drosophila gene smallish (cg43427)

Documents